element 2345 in HTML [4]. 2347 7.1.13 URI 2349 The URI-header field may contain one or more Universal Resource 2350 Identifiers (URIs) by which the resource origin of the entity can 2351 be identified. There is no guarantee that the resource can be 2352 accessed using the URI(s) specified. This field is required for the 2353 201, 301, and 302 response messages, and may be included in any 2354 message that contains resource metainformation. 2356 URI-header = "URI" ":" 1#( "<" URI ">" [ ";" vary ] ) 2358 vary = "vary" "=" <"> 1#vary-dimension <"> 2359 vary-dimension = "type" | "language" | "version" | "encoding" 2360 | "charset" | "user-agent" | extension-vary 2362 extension-vary = token 2364 Any URI specified in this field can be either absolute or relative 2365 to the URI given in the Request-Line. The URI-header improves upon 2366 the Location header field described in Section 7.1.11. For 2367 backwards compatibility with older clients, servers are encouraged 2368 to include both header fields in 301 and 302 responses. 2370 The URI-header may also be used by a client performing a POST 2371 request to suggest a URI for the new entity. Whether or not the 2372 suggested URI is used is entirely up to the server to decide. In 2373 any case, the server's response must include the actual URI(s) of 2374 the new resource if one is successfully created (status 201). 2376 If a URI refers to a set of variants, then the dimensions of that 2377 variance must be given with a vary parameter. One example is: 2379 URI: <http://info.cern.ch/hypertext/WWW/TheProject.multi>; 2380 vary="type,language" 2382 which indicates that the URI covers a group of entities that vary 2383 in media type and natural language. A request for that URI will 2384 result in a response that depends upon the client's request headers 2385 for Accept and Accept-Language. Similar dimensions exist for the 2386 Accept-Encoding, Accept-Charset, Version, and User-Agent header 2387 fields, as demonstrated in the following example. 2389 URI: <TheProject.ps>;vary="encoding,version", 2390 <TheProject.html>;vary="user-agent,charset,version" 2392 The vary parameter has an important effect on cache management, 2393 particularly for caching proxies which service a diverse set of 2394 user agents. Since the response to one user agent may differ from 2395 the response to a second user agent if the two agents have 2396 differing request profiles, a caching proxy must keep track of the 2397 content metainformation for resources with varying dimensions. 2398 Thus, the vary parameter tells the caching proxy what entity 2399 headers must be part of the key for caching that URI. When the 2400 caching proxy gets a request for that URI, it must forward the 2401 request toward the origin server if the request profile includes a 2402 variant dimension that has not already been cached. 2404 7.1.14 Version 2406 The Version field defines the version tag associated with a 2407 rendition of an evolving entity. Together with the Derived-From 2408 field described in Section 7.1.7, it allows a group of people to 2409 work simultaneously on the creation of a work as an iterative 2410 process. The field should be used to allow evolution of a 2411 particular work along a single path. It should not be used to 2412 indicate derived works or renditions in different representations. 2414 Version = "Version" ":" version-tag 2416 version-tag = token | quoted-string 2418 Examples of the Version field include: 2420 Version: 2.1.2 2421 Version: "Fred 19950116-12:26:48" 2422 Version: 2.5a4-omega7 2424 The version tag should be considered opaque to all parties but the 2425 origin server. A user agent can request a particular version of an 2426 entity by including its tag in a Version header as part of the 2427 request. Similarly, a user agent may suggest a value for the 2428 version of an entity transferred via a PUT or POST request. 2429 However, only the origin server can reliably assign or increment 2430 the version tag of an entity. 2432 7.2 Entity Body 2434 The entity-body (if any) sent with an HTTP/1.0 request or response 2435 is in a format and encoding defined by the Entity-Header fields. 2437 Entity-Body = *OCTET 2439 An entity-body is included with a request message only when the 2440 request method calls for one. This specification defines two 2441 request methods, "POST" and "PUT", that allow an entity-body. In 2442 general, the presence of an entity-body in a request is signaled by 2443 the inclusion of a Content-Length and/or Content-Transfer-Encoding 2444 header field in the request message headers. 2446 Note: Most current implementations of the POST and PUT 2447 methods require a valid Content-Length header field. This 2448 can cause problems for some systems that do not know the 2449 size of the entity-body before transmission. Experimental 2450 implementations (and future versions of HTTP) may use a 2451 packetized Content-Transfer-Encoding to obviate the need for 2452 a Content-Length. 2454 For response messages, whether or not an entity-body is included 2455 with a message is dependent on both the request method and the 2456 response code. All responses to the HEAD request method must not 2457 include a body, even though the presence of Content header fields 2458 may lead one to believe they should. Similarly, the responses "204 2459 No Content", "304 Not Modified", and "406 None Acceptable" must not 2460 include a body. 2462 7.2.1 Type 2464 When an Entity-Body is included with a message, the data type of 2465 that body is determined via the header fields Content-Type, 2466 Content-Encoding, and Content-Transfer-Encoding. These define a 2467 three-layer, ordered encoding model: 2469 entity-body <- 2470 Content-Transfer-Encoding( Content-Encoding( Content-Type ) ) 2472 The default for both encodings is none (i.e., the identity 2473 function). A Content-Type specifies the media type of the 2474 underlying data. A Content-Encoding may be used to indicate an 2475 additional encoding mechanism applied to the type (usually for the 2476 purpose of data compression) that is a property of the resource 2477 requested. A Content-Transfer-Encoding may be applied by a 2478 transport agent to ensure safe and/or proper transfer of the 2479 message. Note that the Content-Transfer-Encoding is a property of 2480 the message, not of the resource. 2482 The Content-Type header field has no default value. If and only if 2483 the media type is not given by a Content-Type header (as is always 2484 the case for Simple-Response messages), the receiver may attempt to 2485 guess the media type via inspection of its content and/or the name 2486 extension(s) of the URL used to access the resource. If the media 2487 type remains unknown, the receiver should treat it as type 2488 "application/octet-stream". 2490 7.2.2 Length 2492 When an Entity-Body is included with a message, the length of that 2493 body may be determined in one of several ways. If a Content-Length 2494 header field is present, its value in bytes (number of octets) 2495 represents the length of the Entity-Body. Otherwise, the body 2496 length is determined by the Content-Type (for types with an 2497 explicit end-of-body delimiter), the Content-Transfer-Encoding (for 2498 packetized encodings), or the closing of the connection by the 2499 server. Note that the latter cannot be used to indicate the end of 2500 a request body, since it leaves no possibility for the server to 2501 send back a response. 2503 Note: Some older servers supply an invalid Content-Length 2504 when sending a document that contains server-side includes 2505 dynamically inserted into the data stream. It must be 2506 emphasized that this will not be tolerated by future 2507 versions of HTTP. Unless the client knows that it is 2508 receiving a response from a compliant server, it should not 2509 depend on the Content-Length value being correct. 2511 8. Content Parameters 2513 8.1 Media Types 2515 HTTP uses Internet Media Types [15], formerly referred to as MIME 2516 Content-Types [6], in order to provide open and extensible data 2517 typing and type negotiation. For mail applications, where there is 2518 no type negotiation between sender and receiver, it is reasonable 2519 to put strict limits on the set of allowed media types. With HTTP, 2520 however, user agents can identify acceptable media types as part of 2521 the connection, and thus are allowed more freedom in the use of non- 2522 registered types. The following grammar for media types is a 2523 superset of that for MIME because it does not restrict itself to 2524 the official IANA and x-token types. 2526 media-type = type "/" subtype *( ";" parameter ) 2527 type = token 2528 subtype = token 2530 Parameters may follow the type/subtype in the form of 2531 attribute/value pairs. 2533 parameter = attribute "=" value 2534 attribute = token 2535 value = token | quoted-string 2537 The type, subtype, and parameter attribute names are not case- 2538 sensitive. Parameter values may or may not be case-sensitive, 2539 depending on the semantics of the parameter name. No LWS is allowed 2540 between the type and subtype, nor between an attribute and its 2541 value. 2543 If a given media-type value has been registered by the IANA, any 2544 use of that value must be indicative of the registered data format. 2545 Although HTTP allows the use of non-registered media types, such 2546 usage must not conflict with the IANA registry. Data providers are 2547 strongly encouraged to register their media types with IANA via the 2548 procedures outlined in RFC 1590 [15]. 2550 All media-type's registered by IANA must be preferred over 2551 extension tokens. However, HTTP does not limit conforming 2552 applications to the use of officially registered media types, nor 2553 does it encourage the use of an "x-" prefix for unofficial types 2554 outside of explicitly short experimental use between consenting 2555 applications. 2557 8.1.1 Canonicalization and Text Defaults 2559 Media types are registered in a canonical form. In general, entity 2560 bodies transferred via HTTP must be represented in the appropriate 2561 canonical form prior to transmission. If the body has been encoded 2562 via a Content-Encoding and/or Content-Transfer-Encoding, the data 2563 must be in canonical form prior to that encoding. However, HTTP 2564 modifies the canonical form requirements for media of primary type 2565 "text" and for "application" types consisting of text-like records. 2567 HTTP redefines the canonical form of text media to allow multiple 2568 octet sequences to indicate a text line break. In addition to the 2569 preferred form of CRLF, HTTP applications must accept a bare CR or 2570 LF alone as representing a single line break in text media. 2571 Furthermore, if the text media is represented in a character set 2572 which does not use octets 13 and 10 for CR and LF respectively (as 2573 is the case for some multi-byte character sets), HTTP allows the 2574 use of whatever octet sequence(s) is defined by that character set 2575 to represent the equivalent of CRLF, bare CR, and bare LF. It is 2576 assumed that any recipient capable of using such a character set 2577 will know the appropriate octet sequence for representing line 2578 breaks within that character set. 2580 Note: This interpretation of line breaks applies only to the 2581 contents of an Entity-Body and only after any Content- 2582 Transfer-Encoding and/or Content-Encoding has been removed. 2583 All other HTTP constructs use CRLF exclusively to indicate a 2584 line break. Encoding mechanisms define their own line break 2585 requirements. 2587 A recipient of an HTTP text entity should translate the received 2588 entity line breaks to the local line break conventions before 2589 saving the entity external to the application and its cache; 2590 whether this translation takes place immediately upon receipt of 2591 the entity, or only when prompted by the user, is entirely up to 2592 the individual application. 2594 HTTP also redefines the default character set for text media in an 2595 entity body. If a textual media type defines a charset parameter 2596 with a registered default value of "US-ASCII", HTTP changes the 2597 default to be "ISO-8859-1". Since the character set ISO-8859-1 [19] 2598 is a superset of USASCII [18], this has no effect upon the 2599 interpretation of entity bodies which only contain octets within 2600 the US-ASCII set (0 - 127). The presence of a charset parameter 2601 value in a Content-Type header field overrides the default. 2603 HTTP does not require that the character set of an entity body be 2604 labelled as the lowest common denominator of the character codes 2605 used within a document. 2607 8.1.2 Multipart Types 2609 MIME provides for a number of "multipart" types -- encapsulations of 2610 several entities within a single message's Entity-Body. The 2611 multipart types registered by IANA [17] do not have any special 2612 meaning for HTTP/1.0, though user agents may need to understand 2613 each type in order to correctly interpret the purpose of each body- 2614 part. Ideally, an HTTP user agent should follow the same or similar 2615 behavior as a MIME user agent does upon receipt of a multipart type. 2617 As in MIME [6], all multipart types share a common syntax and must 2618 include a boundary parameter as part of the media type value. The 2619 message body is itself a protocol element and must therefore use 2620 only CRLF to represent line breaks between body-parts. Unlike in 2621 MIME, multipart body-parts may contain HTTP header fields which are 2622 significant to the meaning of that part. 2624 A URI-header field (Section 7.1.13) should be included in the body- 2625 part for each enclosed entity that can be identified by a URI. 2627 8.2 Language Tags 2629 A language tag identifies a natural language spoken, written, or 2630 otherwise conveyed by human beings for communication of information 2631 to other human beings. Computer languages are explicitly excluded. 2632 The HTTP/1.0 protocol uses language tags within the Accept-Language 2633 and Content-Language header fields. 2635 The syntax and registry of HTTP language tags is the same as that 2636 defined by RFC 1766 [1]. In summary, a language tag is composed of 2637 1 or more parts: A primary language tag and a (possibly empty) 2638 series of subtags: 2640 language-tag = primary-tag *( "-" subtag ) 2642 primary-tag = 1*8ALPHA 2643 subtag = 1*8ALPHA 2645 Whitespace is not allowed within the tag and all tags are to be 2646 treated as case insensitive. The namespace of language tags is 2647 administered by the IANA. Example tags include: 2649 en, en-US, en-cockney, i-cherokee, x-pig-latin 2651 where any two-letter primary-tag is an ISO 639 language 2652 abbreviation and any two-letter initial subtag is an ISO 3166 2653 country code. 2655 Note: Earlier versions of this document specified an 2656 incomplete language tag, where values were limited to ISO 2657 639 language abbreviations with an optional ISO 3166 country 2658 code appended after an underscore ("_") or slash ("/") 2659 character. This format was abandoned in favor of the 2660 recently proposed standard for Internet protocols. 2662 8.3 Character Sets 2664 HTTP uses the same definition of the term "character set" as that 2665 described for MIME: 2667 The term "character set" is used in this document to 2668 refer to a method used with one or more tables to convert 2669 a sequence of octets into a sequence of characters. Note 2670 that unconditional conversion in the other direction is 2671 not required, in that not all characters may be available 2672 in a given character set and a character set may provide 2673 more than one sequence of octets to represent a 2674 particular character. This definition is intended to 2675 allow various kinds of character encodings, from simple 2676 single-table mappings such as US-ASCII to complex table 2677 switching methods such as those that use ISO 2022's 2678 techniques. However, the definition associated with a 2679 MIME character set name must fully specify the mapping to 2680 be performed from octets to characters. In particular, 2681 use of external profiling information to determine the 2682 exact mapping is not permitted. 2684 Character sets are identified by case-insensitive tokens. The 2685 complete set of allowed charset values are defined by the IANA 2686 Character Set registry [17]. However, because that registry does 2687 not define a single, consistent token for each character set, we 2688 define here the preferred names for those character sets most 2689 likely to be used with HTTP entities. This set of charset values 2690 includes those registered by RFC 1521 [6] -- the US-ASCII [18] and 2691 ISO8859 [19] character sets -- and other character set names 2692 specifically recommended for use within MIME charset parameters. 2694 charset = "US-ASCII" 2695 | "ISO-8859-1" | "ISO-8859-2" | "ISO-8859-3" 2696 | "ISO-8859-4" | "ISO-8859-5" | "ISO-8859-6" 2697 | "ISO-8859-7" | "ISO-8859-8" | "ISO-8859-9" 2698 | "ISO-2022-JP" | "ISO-2022-JP-2" | "ISO-2022-KR" 2699 | "UNICODE-1-1" | "UNICODE-1-1-UTF-7" | "UNICODE-1-1-UTF-8" 2700 | token 2702 Although HTTP allows an arbitrary token to be used as a character 2703 set value, any token that has a predefined value within the IANA 2704 Character Set registry [17] must represent the character set 2705 defined by that registry. Applications are encouraged, but not 2706 required, to limit their use of character sets to those defined by 2707 the IANA registry. 2709 8.4 Encoding Mechanisms 2711 Encoding mechanism values are used to indicate an encoding 2712 transformation that has been or can be applied to a resource. 2713 Encoding mechanisms are primarily used to allow a document to be 2714 compressed or encrypted without losing the identity of its 2715 underlying media type. Typically, the resource is stored with this 2716 encoding and is only decoded before rendering or analogous usage. 2718 encoding-mechanism = "gzip" | "compress" | token 2720 Note: For historical reasons, HTTP/1.0 applications should 2721 consider "x-gzip" and "x-compress" to be equivalent to 2722 "gzip" and "compress", respectively. 2724 All encoding-mechanism values are case-insensitive. HTTP/1.0 uses 2725 encoding-mechanism values in the Accept-Encoding (Section 5.4.3) 2726 and Content-Encoding (Section 7.1.2) header fields. Although the 2727 value describes the encoding-mechanism, what is more important is 2728 that it indicates what decoding mechanism will be required to 2729 remove the encoding. Note that a single program may be capable of 2730 decoding multiple encoding-mechanism formats. Two values are 2731 defined by this specification: 2733 gzip An encoding format produced by the file compression program 2734 "gzip" (GNU zip) developed by Jean-loup Gailly. This format 2735 is typically a Lempel-Ziv coding (LZ77) with a 32 bit CRC. 2736 Gzip is available from the GNU project at 2737 <URL:ftp://prep.ai.mit.edu/pub/gnu/>. 2739 compress The encoding format produced by the file compression 2740 program "compress". This format is an adaptive Lempel-Ziv- 2741 Welch coding (LZW). 2743 8.5 Transfer Encodings 2745 Transfer encoding values are used to indicate an encoding 2746 transformation that has been, can be, or may need to be applied to 2747 an Entity-Body in order to ensure safe transport through the 2748 network. Transfer encodings are only used with entities destined 2749 for or retrieved from MIME-conformant systems, and thus will rarely 2750 occur in an HTTP/1.0 message. This differs from an encoding- 2751 mechanism in that the transfer encoding is a property of the 2752 message, not of the original resource. 2754 transfer-encoding = "binary" | "8bit" | "7bit" 2755 | "quoted-printable" | "base64" 2756 | token 2758 All transfer-encoding values are case-insensitive. HTTP/1.0 uses 2759 transfer-encoding values in the Accept-Encoding (Section 5.4.3) and 2760 Content-Transfer-Encoding (Section 7.1.5) header fields. 2762 The values "7bit", "8bit", and "binary" are used to indicate that 2763 no transfer encoding has been performed. Instead, they describe the 2764 sort of encoding that might be needed for transmission through an 2765 unsafe transport system. Binary indicates that the body may contain 2766 any set of octets. 8bit adds the restrictions that CR and LF 2767 characters only occur as part of CRLF line separators, all lines 2768 are short (less than 1000 octets), and no NULs (octet 0) are 2769 present. 7bit adds a further restriction that all octets are 7-bit 2770 US-ASCII characters. 2772 The "quoted-printable" and "base64" values indicate that the 2773 associated encoding (as defined in MIME [6]) has been applied to 2774 the body. These encodings consist entirely of 7-bit US-ASCII 2775 characters. 2777 9. Content Negotiation 2779 Content negotiation is an optional feature of the HTTP protocol. It 2780 is designed to allow for preemptive selection of a preferred 2781 content representation, within a single request-response round-trip, 2782 and without intervention from the user. However, this may not 2783 always be desirable for the user and is sometimes unnecessary for 2784 the content provider. Implementors are encouraged to provide 2785 mechanisms whereby the amount of preemptive content negotiation, 2786 and the parameters of that negotiation, are configurable by the 2787 user and server maintainer. 2789 The first step in the negotiation algorithm is for the server to 2790 determine whether or not there are any content variants for the 2791 requested resource. Content variants may be in the form of multiple 2792 preexisting entities or a set of dynamic conversion filters. If 2793 there are no variant forms of the resource, the "negotiation" is 2794 limited to whether or not that single media type is acceptable 2795 under the constraints given by the Accept request header field 2796 (if any). 2798 If variants are available, those variants that are completely 2799 unacceptable should be removed from consideration first. 2800 Unacceptable variants include those with a Content-Encoding not 2801 listed in an Accept-Encoding field, those with a character subset 2802 (other than the default ISO-8859-1) not listed in an Accept-Charset 2803 field, and those with a media type not within any of the media 2804 ranges of an explicitly constrained Accept field (or listed with a 2805 zero quality parameter). 2807 If no acceptable variants remain at this point, the server should 2808 respond with a "406 None Acceptable" response message. If more than 2809 one variant remains, and at least one has a Content-Language within 2810 those listed by an Accept-Language field, any variants which do not 2811 match the language constraint are removed from further 2812 consideration. 2814 If multiple choices still remain, the selection is further winnowed 2815 by calculating and comparing the relative quality of the available 2816 media types. The calculated weights are normalized to a real number 2817 in the range 0 through 1, where 0 is the minimum and 1 the maximum 2818 value. The following parameters are included in the calculation: 2820 q The quality factor chosen by the user agent (and 2821 configurable by the user) that represents the desirability 2822 of that media type. In this case, desirability is usually a 2823 measure of the clients ability to faithfully represent the 2824 contents of that media type to the user. The value is in 2825 the range [0,1], where the default value is 1. 2827 qs The quality factor, as chosen by the server (via some 2828 unspecified mechanism), to represent the relative quality 2829 of a particular variant representation of the source. For 2830 example, a picture originally in JPEG form would have a 2831 lower qs when translated to the XBM format, and much lower 2832 qs when translated to an ASCII-art representation. Note, 2833 however, that this is a function of the source -- an 2834 original piece of ASCII-art may degrade in quality if it is 2835 captured in JPEG form. The qs value has the same range and 2836 default as q. 2838 mxb The maximum number of bytes in the Entity-Body accepted by 2839 the client. The default value is mxb=undefined 2840 (i.e. infinity). 2842 bs The actual number of bytes in the Entity-Body for a 2843 particular source representation. This should equal the 2844 value sent as Content-Length. The default value is 0. 2846 The discrete mapping function is defined as: 2848 { if mxb=undefined, then (qs * q) } 2849 Q(q,qs,mxb,bs) = { if mxb >= bs, then (qs * q) } 2850 { if mxb < bs, then 0 } 2852 The variants with a maximal value for the Q function represent the 2853 preferred representation(s) of the entity. If multiple 2854 representations exist for a single media type (as would be the case 2855 if they only varied by Content-Encoding), then the smallest 2856 representation (lowest bs) is preferred. 2858 Finally, there may still be multiple choices available to the user. 2859 If so, the server may either choose one (possibly at random) from 2860 those available and respond with "200 OK", or it may respond with 2861 "300 Multiple Choices" and include an entity describing the 2862 choices. In the latter case, the entity should either be of type 2863 "text/html' (such that the user can choose from among the choices 2864 by following an exact link) or of some type that would allow the 2865 user agent to perform the selection automatically (no such type is 2866 available at the time of this writing). 2868 The "300 Multiple Choices" response can be given even if the server 2869 does not perform any winnowing of the representation choices via 2870 the content negotiation algorithm described above. Furthermore, it 2871 may include choices that were not considered as part of the 2872 negotiation algorithm and resources that may be located at other 2873 servers. 2875 10. Access Authentication 2877 HTTP provides a simple challenge-response authorization mechanism 2878 which may be used by a server to challenge a client request and by 2879 a client to provide authentication information. The mechanism uses 2880 an extensible token to identify the authentication scheme, followed 2881 by a comma-separated list of attribute-value pairs which carry the 2882 parameters necessary for achieving authentication via that scheme. 2884 auth-scheme = "Basic" | token 2886 auth-param = token "=" quoted-string 2888 The "401 Unauthorized" response message is used by an origin server 2889 to challenge the authorization of a user agent. This response must 2890 include a WWW-Authenticate header field containing the challenge 2891 applicable to the requested resource. 2893 challenge = auth-scheme 1*LWS realm [ "," 1#auth-param ] 2895 realm = "Realm" "=" quoted-string 2897 The realm attribute is required for all access authentication 2898 schemes which issue a challenge. The realm value, in combination 2899 with the root URL of the server being accessed, defines the 2900 authorization space. These realms allow the protected resources on 2901 a server to be partitioned into a set of authorization spaces, each 2902 with its own authentication scheme and/or database. The realm value 2903 is a string, generally assigned by the origin server, which may 2904 have additional semantics specific to the authentication scheme. 2906 A user agent that wishes to authenticate itself with a server 2907 (usually, but not necessarily, after receiving a 401 response), may 2908 do so by including an Authorization header field with the request. 2909 The Authorization field value consists of credentials containing 2910 the authentication information of the user agent for the realm of 2911 the resource being requested. 2913 credentials = auth-scheme [ 1*LWS encoded-cookie ] #auth-param 2915 encoded-cookie = 1*<any CHAR except CTLs or tspecials, 2916 but including "=" and "/"> 2918 The domain over which credentials can be automatically applied by a 2919 user agent is determined by the authorization space. If a request 2920 is authenticated, the credentials can be reused for all other 2921 requests within that authorization space for a period of time 2922 determined by the authentication scheme, parameters, and/or user 2923 preference. 2925 The HTTP protocol does not restrict applications to this simple 2926 challenge-response mechanism for access authentication. Additional 2927 mechanisms may be used at the transport level, via message 2928 encapsulation, and/or with additional header fields specifying 2929 authentication information. However, these additional mechanisms 2930 are not defined by this specification. 2932 Proxies must be completely transparent regarding user agent access 2933 authentication. That is, they must forward the WWW-Authenticate and 2934 Authorization headers untouched. HTTP/1.0 does not provide a means 2935 for a client to be authenticated with a proxy -- this feature will 2936 be available in future versions of HTTP. 2938 Note: The names Proxy-Authenticate and Proxy-Authorization 2939 have been suggested as headers, analogous to WWW-Authenticate 2940 and Authorization, but applying only to the immediate 2941 connection with a proxy. 2943 10.1 Basic Authentication Scheme 2945 The basic authentication scheme is based on the model that the 2946 client must authenticate itself with a user-ID and a password for 2947 each realm. The realm value should be considered an opaque string 2948 which can only be compared for equality with other realms. The 2949 server will service the request only if it can validate the user-ID 2950 and password for the domain of the Request-URI. 2952 basic-challenge = "Basic" SP realm 2954 The client sends the user-ID and password (separated by a single 2955 colon ":" character) within a base64 [6] encoded-cookie in the 2956 credentials. 2958 basic-credentials = "Basic" SP basic-cookie 2959 basic-cookie = <base64 encoding of userid-password> 2960 userid-password = [ token ] ":" *text 2962 There are no optional authentication parameters for the basic 2963 scheme. For example, if the user agent wishes to send the user-ID 2964 "Aladdin" and password "open sesame", it would use the following 2965 header field: 2967 Authorization: Basic QWxhZGRpbjpvcGVuIHNlc2FtZQ== 2969 The basic authentication scheme is a non-secure method of filtering 2970 unauthorized access to resources on an HTTP server. It is based on 2971 the assumption that the connection between the client and the 2972 server can be regarded as a trusted carrier. As this is not 2973 generally true on an open network, the basic authentication scheme 2974 should be used accordingly. In spite of this, clients are strongly 2975 encouraged to implement the scheme in order to communicate with 2976 servers that use it. 2978 11. Security Considerations 2980 This section is meant to inform application developers, information 2981 providers, and users of the security limitations in HTTP/1.0 as 2982 described by this document. The discussion does not include 2983 definitive solutions to the problems revealed, though it does make 2984 some suggestions for reducing security risks. 2986 11.1 Authentication of Clients 2988 As mentioned in Section 10.1, the Basic authentication scheme is 2989 not considered to be a secure method of user authentication, nor 2990 does it prevent the Entity-Body from being transmitted in clear 2991 text across the physical network used as the carrier. HTTP/1.0 does 2992 not prevent additional authentication schemes and encryption 2993 mechanisms to be employed to increase security. 2995 11.2 Idempotent Methods 2997 The writers of client software should be aware that the software 2998 represents the user in their interactions over the net, and should 2999 be careful to allow the user to be aware of any actions they may 3000 take which may have an unexpected significance to themselves or 3001 others. 3003 In particular, the convention has been established that the GET and 3004 HEAD methods should never have the significance of taking an 3005 action. The link "click here to subscribe"--causing the reading of a 3006 special "magic" document--is open to abuse by others making a link 3007 "click here to see a pretty picture." These methods should be 3008 considered "safe" and should not have side effects. This allows the 3009 client software to represent other methods (such as POST, PUT and 3010 DELETE) in a special way, so that the user is aware of the fact 3011 that an action is being requested. 3013 11.3 Abuse of Server Log Information 3015 A server is in the position to save personal data about a user's 3016 requests which may identify their reading patterns or subjects of 3017 interest. This information is clearly confidential in nature and 3018 its handling may be constrained by law in certain countries. People 3019 using the HTTP protocol to provide data are responsible for 3020 ensuring that such material is not distributed without the 3021 permission of any individuals that are identifiable by the 3022 published results. 3024 Two header fields are worth special mention in this context: 3025 Referer and From. The Referer field allows reading patterns to be 3026 studied and reverse links drawn. Although it can be very useful, 3027 its power can be abused if user details are not separated from the 3028 information contained in the Referer. Even when the personal 3029 information has been removed, the Referer field may have indicated 3030 a secure document's URI, whose revelation itself would be a breach 3031 of security. 3033 The information sent in the From field might conflict with the 3034 user's privacy interests or their site's security policy, and hence 3035 it should not be transmitted without the user being able to 3036 disable, enable, and modify the contents of the field. The user 3037 must be able to set the active contents of this field within a user 3038 preference or application defaults configuration. 3040 We suggest, though do not require, that a convenient toggle 3041 interface be provided for the user to enable or disable the sending 3042 of From and Referer information. 3044 12. Acknowledgments 3046 This specification makes heavy use of the augmented BNF and generic 3047 constructs defined by David H. Crocker for RFC 822 [8]. Similarly, 3048 it reuses many of the definitions provided by Nathaniel Borenstein 3049 and Ned Freed for MIME [6]. We hope that their inclusion in this 3050 specification will help reduce past confusion over the relationship 3051 between HTTP/1.0 and Internet mail. 3053 The HTTP protocol has evolved considerably over the past three 3054 years. It has benefited from a large and active developer community-- 3055 the many people who have participated on the www-talk mailing list-- 3056 and it is that community which has been most responsible for the 3057 success of HTTP and of the World-Wide Web in general. 3058 Marc Andreessen, Robert Cailliau, Daniel W. Connolly, Bob Denny, 3059 Phillip M. Hallam-Baker, Haringkon W. Lie, Ari Luotonen, Rob McCool, 3060 Dave Raggett, Tony Sanders, and Marc VanHeyningen deserve special 3061 recognition for their efforts in defining aspects of the protocol 3062 for early versions of this specification. 3064 This document has benefited greatly from the comments of all those 3065 participating in the HTTPWG. In addition to those already 3066 mentioned, the following individuals have contributed to this 3067 specification: 3069 Gary Adams Harald Tveit Alvestrand 3070 Keith Ball Brian Behlendorf 3071 Paul Burchard Maurizio Codogno 3072 Mike Cowlishaw Michael A. Dolan 3073 John Franks Alex Hopmann 3074 Bob Jernigan Martijn Koster 3075 Dave Kristol Daniel LaLiberte 3076 Albert Lunde John C. Mallery 3077 Larry Masinter Mitra 3078 Gavin Nicol Marc Salomon 3079 Chuck Shotton Eric W. Sink 3080 Simon E. Spero 3082 13. References 3084 [1] H. T. Alvestrand. "Tags for the identification of languages." 3085 RFC 1766, UNINETT, March 1995. 3087 [2] F. Anklesaria, M. McCahill, P. Lindner, D. Johnson, D. Torrey, 3088 and B. Alberti. "The Internet Gopher Protocol: A distributed 3089 document search and retrieval protocol." RFC 1436, University 3090 of Minnesota, March 1993. 3092 [3] T. Berners-Lee. "Universal Resource Identifiers in WWW: A 3093 Unifying Syntax for the Expression of Names and Addresses of 3094 Objects on the Network as used in the World-Wide Web." 3095 RFC 1630, CERN, June 1994. 3097 [4] T. Berners-Lee and D. Connolly. "HyperText Markup Language 3098 Specification - 2.0." Work in Progress (draft-ietf-html-spec- 3099 01.txt), CERN, HaL Computer Systems, February 1995. 3101 [5] T. Berners-Lee, L. Masinter, and M. McCahill. "Uniform Resource 3102 Locators (URL)." RFC 1738, CERN, Xerox PARC, University of 3103 Minnesota, October 1994. 3105 [6] N. Borenstein and N. Freed. "MIME (Multipurpose Internet Mail 3106 Extensions) Part One: Mechanisms for Specifying and Describing 3107 the Format of Internet Message Bodies." RFC 1521, Bellcore, 3108 Innosoft, September 1993. 3110 [7] R. Braden. "Requirements for Internet hosts - application and 3111 support." STD 3, RFC 1123, IETF, October 1989. 3113 [8] D. H. Crocker. "Standard for the Format of ARPA Internet Text 3114 Messages." STD 11, RFC 822, UDEL, August 1982. 3116 [9] F. Davis, B. Kahle, H. Morris, J. Salem, T. Shen, R. Wang, 3117 J. Sui, and M. Grinbaum. "WAIS Interface Protocol Prototype 3118 Functional Specification." (v1.5), Thinking Machines 3119 Corporation, April 1990. 3121 [10] R. T. Fielding. "Relative Uniform Resource Locators." Work in 3122 Progress (draft-ietf-uri-relative-url-05.txt), UC Irvine, 3123 January 1995. 3125 [11] M. Horton and R. Adams. "Standard for interchange of USENET 3126 messages." RFC 1036 (Obsoletes RFC 850), AT&T Bell 3127 Laboratories, Center for Seismic Studies, December 1987. 3129 [12] B. Kantor and P. Lapsley. "Network News Transfer Protocol: A 3130 Proposed Standard for the Stream-Based Transmission of News." 3131 RFC 977, UC San Diego, UC Berkeley, February 1986. 3133 [13] K. Moore. "MIME (Multipurpose Internet Mail Extensions) Part 3134 Two: Message Header Extensions for Non-ASCII Text." RFC 1522, 3135 University of Tennessee, September 1993. 3137 [14] J. Postel. "Simple Mail Transfer Protocol." STD 10, RFC 821, 3138 USC/ISI, August 1982. 3140 [15] J. Postel. "Media Type Registration Procedure." RFC 1590, 3141 USC/ISI, March 1994. 3143 [16] J. Postel and J. K. Reynolds. "File Transfer Protocol (FTP)." 3144 STD 9, RFC 959, USC/ISI, October 1985. 3146 [17] J. Reynolds and J. Postel. "Assigned Numbers." STD 2, RFC 1700, 3147 USC/ISI, October 1994. 3149 [18] US-ASCII. Coded Character Set - 7-Bit American Standard Code for 3150 Information Interchange. Standard ANSI X3.4-1986, ANSI, 1986. 3152 [19] ISO-8859. International Standard -- Information Processing -- 3153 8-bit Single-Byte Coded Graphic Character Sets -- Part 1: Latin 3154 Alphabet No. 1, ISO 8859-1:1987. Part 2: Latin alphabet No. 2, 3155 ISO 8859-2, 1987. Part 3: Latin alphabet No. 3, ISO 8859-3, 3156 1988. Part 4: Latin alphabet No. 4, ISO 8859-4, 1988. Part 5: 3157 Latin/Cyrillic alphabet, ISO 8859-5, 1988. Part 6: Latin/Arabic 3158 alphabet, ISO 8859-6, 1987. Part 7: Latin/Greek alphabet, ISO 3159 8859-7, 1987. Part 8: Latin/Hebrew alphabet, ISO 8859-8, 1988. 3160 Part 9: Latin alphabet No. 5, ISO 8859-9, 1990. 3162 14. Authors' Addresses 3164 Tim Berners-Lee 3165 Director, W3 Consortium 3166 MIT Laboratory for Computer Science 3167 545 Technology Square 3168 Cambridge, MA 02139, U.S.A. 3169 Tel: +1 (617) 253 9670 3170 Fax: +1 (617) 258 8682 3171 Email: timbl@w3.org 3173 Roy T. Fielding 3174 Department of Information and Computer Science 3175 University of California 3176 Irvine, CA 92717-3425, U.S.A. 3177 Tel: +1 (714) 824-4049 3178 Fax: +1 (714) 824-4056 3179 Email: fielding@ics.uci.edu 3181 Henrik Frystyk Nielsen 3182 World-Wide Web Project 3183 CERN, 3184 1211 Geneva 23, Switzerland 3185 Tel: +41 (22) 767 8265 3186 Fax: +41 (22) 767 8730 3187 Email: frystyk@w3.org 3189 Appendices 3191 These appendices are provided for informational reasons only -- they 3192 do not form a part of the HTTP/1.0 specification. 3194 A. Internet Media Type message/http 3196 In addition to defining the HTTP/1.0 protocol, this document serves 3197 as the specification for the Internet media type "message/http". 3198 The following is to be registered with IANA [15]. 3200 Media Type name: message 3202 Media subtype name: http 3204 Required parameters: none 3206 Optional parameters: version, type 3208 version: The HTTP-Version number of the enclosed message 3209 (e.g. "1.0"). If not present, the version can be 3210 determined from the first line of the body. 3212 type: The message type -- "request" or "response". If 3213 not present, the type can be determined from the 3214 first line of the body. 3216 Encoding considerations: only "7bit", "8bit", or "binary" are 3217 permitted 3219 Security considerations: none 3221 B. Minimum Compliance 3223 Early reviews of this specification have indicated the need for a 3224 statement of the minimum requirements for an implementation to be 3225 considered in compliance with HTTP/1.0. This section will be 3226 written soon. 3228 Note: The primary difficulty in determining a standard for 3229 minimum compliance rests in the fact that HTTP is a flexible 3230 protocol which can be used for many purposes. The 3231 requirements for special purpose applications often differ 3232 from those of general purpose applications. 3234 B.1 Servers 3236 Servers have a special responsibility for being honest when 3237 generating their responses to requesting clients. The Status-Code 3238 sent in the Status-Line must correspond to the actual action taken 3239 by the server. This is especially the case when the method used in 3240 the request is one of PUT, POST, DELETE, LINK, and UNLINK. If a 3241 Status-Code of 200 is returned, the client must be able to assume 3242 that the action has been carried out. If the server is not able to 3243 fulfill the requested action immediately, the correct status code 3244 to use is "202 Accepted". 3246 The methods GET and HEAD must be supported by all general-purpose 3247 servers. Servers which provide Last-Modified dates for resources 3248 must also support the conditional GET method. 3250 C. Tolerant Applications 3252 While it may be appropriate for testing applications to verify full 3253 conformance to this specification, it is recommended that 3254 operational applications be tolerant of deviations. This appendix 3255 mentions the most important topics where tolerance is recommended. 3257 C.1 Request-Line, Status-Line, and Header Fields 3259 Clients should be tolerant in parsing the Status-Line and servers 3260 tolerant when parsing the Request-Line. In particular, they should 3261 accept any amount of SP and HTAB characters between fields, even 3262 though only a single SP is specified. 3264 The line terminator for HTTP-header fields should be the sequence 3265 CRLF. However, we recommend that applications, when parsing such 3266 headers, recognize a single LF as a line terminator and ignore the 3267 leading CR. 3269 We recommend that servers allocate URIs free of "variant" 3270 characters (characters whose representation differs in some of the 3271 national variant character sets), punctuation characters, and 3272 spaces. This makes URIs easier to handle by humans when the need 3273 arises (such as for debugging or transmission through non hypertext 3274 systems). 3276 D. Relationship to MIME 3278 HTTP/1.0 reuses many of the constructs defined for Internet Mail 3279 (RFC 822 [8]) and the Multipurpose Internet Mail Extensions 3280 (MIME [6]) to allow entities to be transmitted in an open variety 3281 of representations and with extensible mechanisms. However, HTTP is 3282 not a MIME-conforming application. HTTP's performance requirements 3283 differ substantially from those of Internet mail. Since it is not 3284 limited by the restrictions of existing mail protocols and 3285 gateways, HTTP does not obey some of the constraints imposed by 3286 RFC 822 and MIME for mail transport. 3288 This appendix describes specific areas where HTTP differs from 3289 MIME. Gateways to MIME-compliant protocols must be aware of these 3290 differences and provide the appropriate conversions where 3291 necessary. No conversion should be necessary for a MIME-conforming 3292 entity to be transferred using HTTP. 3294 D.1 Conversion to Canonical Form 3296 MIME requires that an entity be converted to canonical form prior 3297 to being transferred, as described in Appendix G of RFC 1521 [6]. 3298 Although HTTP does require media types to be transferred in 3299 canonical form, it changes the definition of "canonical form" for 3300 text-based media types as described in Section 8.1.1. 3302 D.1.1 Representation of Line Breaks 3304 MIME requires that the canonical form of any text type represent 3305 line breaks as CRLF and forbids the use of CR or LF outside of line 3306 break sequences. Since HTTP allows CRLF, bare CR, and bare LF (or 3307 the octet sequence(s) to which they would be translated for the 3308 given character set) to indicate a line break within text content, 3309 recipients of an HTTP message cannot rely upon receiving MIME- 3310 canonical line breaks in text. 3312 Where it is possible, a gateway from HTTP to a MIME-conformant 3313 protocol should translate all line breaks within text/* media types 3314 to the MIME canonical form of CRLF. However, this may be 3315 complicated by the presence of a Content-Encoding and by the fact 3316 that HTTP allows the use of some character sets which do not use 3317 octets 13 and 10 to represent CR and LF (as is the case for some 3318 multi-byte character sets). 3320 D.1.2 Default Character Set 3322 MIME requires that all subtypes of the top-level Content-Type 3323 "text" have a default character set of US-ASCII [18]. In contrast, 3324 HTTP defines the default character set for "text" to be 3325 ISO88591 [19] (a superset of US-ASCII). Therefore, if a text/* 3326 media type given in the Content-Type header field does not already 3327 include an explicit charset parameter, the parameter 3329 ;charset="iso-8859-1" 3331 should be added by the gateway if the entity contains any octets 3332 greater than 127. 3334 D.2 Default Content-Transfer-Encoding 3336 The default Content-Transfer-Encoding (CTE) for all MIME messages 3337 is "7bit". In contrast, HTTP defines the default CTE to be 3338 "binary". Therefore, if an entity does not include an explicit CTE 3339 header field, the gateway should apply either the "quoted-printable" 3340 or "base64" transfer encodings and add the appropriate 3341 Content-Transfer-Encoding field. At a minimum, the explicit CTE 3342 field of 3344 Content-Transfer-Encoding: binary 3346 should be added by the gateway if it is unwilling to apply a mail- 3347 safe encoding. 3349 D.3 Introduction of Content-Encoding 3351 MIME does not include any concept equivalent to HTTP's Content- 3352 Encoding header field. Since this acts as a modifier on the media 3353 type, gateways to MIME-conformant protocols should either change 3354 the value of the Content-Type header field or decode the Entity- 3355 Body before forwarding the message. 3357 Note: Some experimental applications of Content-Type for 3358 Internet mail have used a media-type parameter of 3359 ";conversions=<encoding-mechanisms>" to perform an 3360 equivalent function as Content-Encoding. However, this 3361 parameter is not part of the MIME specification at the time 3362 of this writing. 3364 E. Example of Version Control 3366 This appendix gives an example on how the Entity-Header fields 3367 Version and Derived-From can be used to apply version control to 3368 the creation and parallel development of a work. In order to 3369 simplify the example, only two user agents (A and B) are considered 3370 together with an origin server S. 3372 o A sends a POST request to S, including the header 3373 "Version: 1.0" and an entity 3375 o S replies "201 Created" to A, including the header 3376 "Version: 1.0" and a URI-header which should be used for 3377 future references 3379 o A starts editing the entity 3381 o B sends a GET request to S 3383 o S replies "200 OK" to B, including the entity with a header 3384 "Version: 1.0" 3386 o B starts editing the entity 3388 o B sends a PUT request to S, including the entity and a header 3389 "Derived-From: 1.0" 3391 o S replies "204 No Content" to B, including a header 3392 "Version: 1.1" but no entity 3394 o A sends a PUT request to S, including the entity and a header 3395 "Derived-From: 1.0" 3397 o S replies "409 Conflict" to A, including "Version: 1.1" and the 3398 list of problems with merging A's changes to 1.0 with those 3399 already applied for B and version 1.1 3401 o A merges B's changes with its own, possibly with help from the 3402 user of A 3404 o A sends a PUT request to S including the entity and the header 3405 "Derived-From: 1.1" 3407 o S replies "204 No Content" to A, including the header 3408 "Version: 1.2" but no entity 3410 The example can be expanded to any number of involved user agents, 3411 though the likelihood of conflicts and the difficulty of resolving 3412 them may increase.

idnits 2.17.1 draft-ietf-http-v10-spec-00.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** Cannot find the required boilerplate sections (Copyright, IPR, etc.) in this document. Expected boilerplate is as follows today (2024-04-25) according to https://trustee.ietf.org/license-info : IETF Trust Legal Provisions of 28-dec-2009, Section 6.a: This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79. IETF Trust Legal Provisions of 28-dec-2009, Section 6.b(i), paragraph 2: Copyright (c) 2024 IETF Trust and the persons identified as the document authors. All rights reserved. IETF Trust Legal Provisions of 28-dec-2009, Section 6.b(i), paragraph 3: This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- ** Missing expiration date. The document expiration date should appear on the first and last page. ** The document seems to lack a 1id_guidelines paragraph about Internet-Drafts being working documents. ** The document seems to lack a 1id_guidelines paragraph about the list of current Internet-Drafts. ** The document seems to lack a 1id_guidelines paragraph about the list of Shadow Directories. == No 'Intended status' indicated for this document; assuming Proposed Standard == The page length should not exceed 58 lines per page, but there was 1 longer page, the longest (page 1) being 3417 lines Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- ** The document seems to lack an IANA Considerations section. (See Section 2.2 of https://www.ietf.org/id-info/checklist for how to handle the case when there are no actions for IANA.) ** The document seems to lack separate sections for Informative/Normative References. All references will be assumed normative when checking for downward references. == There are 5 instances of lines with non-RFC2606-compliant FQDNs in the document. Miscellaneous warnings: ---------------------------------------------------------------------------- -- The document seems to lack a disclaimer for pre-RFC5378 work, but may have content which was first submitted before 10 November 2008. If you have contacted all the original authors and they are all willing to grant the BCP78 rights to the IETF Trust, then this is fine, and you can ignore this comment. If not, you may need to add the pre-RFC5378 disclaimer. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (March 8, 1995) is 10641 days in the past. Is this intentional? -- Found something which looks like a code comment -- if you have code sections in the document, please surround them with '

' and
     '' lines.


  Checking references for intended status: Proposed Standard
  ----------------------------------------------------------------------------

     (See RFCs 3967 and 4897 for information about using normative references
     to lower-maturity documents in RFCs)

  -- Looks like a reference, but probably isn't: 'CRLF' on line 411

  == Missing Reference: '0' is mentioned on line 2825, but not defined

  == Unused Reference: '10' is defined on line 3121, but no explicit
     reference was found in the text

  ** Obsolete normative reference: RFC 1766 (ref. '1') (Obsoleted by RFC
     3066, RFC 3282)

  ** Downref: Normative reference to an Informational RFC: RFC 1436 (ref. '2')

  ** Downref: Normative reference to an Informational RFC: RFC 1630 (ref. '3')

  == Outdated reference: A later version (-05) exists of
     draft-ietf-html-spec-01

  ** Downref: Normative reference to an Historic draft: draft-ietf-html-spec
     (ref. '4')

  ** Obsolete normative reference: RFC 1738 (ref. '5') (Obsoleted by RFC
     4248, RFC 4266)

  ** Obsolete normative reference: RFC 1521 (ref. '6') (Obsoleted by RFC
     2045, RFC 2046, RFC 2047, RFC 2048, RFC 2049)

  ** Obsolete normative reference: RFC  822 (ref. '8') (Obsoleted by RFC 2822)

  -- Possible downref: Non-RFC (?) normative reference: ref. '9'

  ** Obsolete normative reference: RFC  850 (ref. '11') (Obsoleted by RFC
     1036)

  ** Obsolete normative reference: RFC  977 (ref. '12') (Obsoleted by RFC
     3977)

  ** Obsolete normative reference: RFC 1522 (ref. '13') (Obsoleted by RFC
     2045, RFC 2046, RFC 2047, RFC 2048, RFC 2049)

  ** Obsolete normative reference: RFC  821 (ref. '14') (Obsoleted by RFC
     2821)

  ** Obsolete normative reference: RFC 1590 (ref. '15') (Obsoleted by RFC
     2045, RFC 2046, RFC 2047, RFC 2048, RFC 2049)

  ** Obsolete normative reference: RFC 1700 (ref. '17') (Obsoleted by RFC
     3232)

  -- Possible downref: Non-RFC (?) normative reference: ref. '18'

  -- Possible downref: Non-RFC (?) normative reference: ref. '19'


     Summary: 20 errors (**), 0 flaws (~~), 6 warnings (==), 7 comments (--).

     Run idnits with the --verbose option for more detailed information about
     the items above.

--------------------------------------------------------------------------------

1	HTTP Working Group                                        T. Berners-Lee
2	INTERNET-DRAFT                                            R. T. Fielding
3	                     H. Frystyk Nielsen
4	Expires September 8, 1995                                  March 8, 1995

6	                Hypertext Transfer Protocol -- HTTP/1.0

8	Status of this Memo

10	   This document is an Internet-Draft. Internet-Drafts are working
11	   documents of the Internet Engineering Task Force (IETF), its areas,
12	   and its working groups. Note that other groups may also distribute
13	   working documents as Internet-Drafts.

15	   Internet-Drafts are draft documents valid for a maximum of six
16	   months and may be updated, replaced, or obsoleted by other
17	   documents at any time. It is inappropriate to use Internet-Drafts
18	   as reference material or to cite them other than as "work in
19	   progress".

21	   To learn the current status of any Internet-Draft, please check the
22	   "1id-abstracts.txt" listing contained in the Internet-Drafts Shadow
23	   Directories on ftp.is.co.za (Africa), nic.nordu.net (Europe),
24	   munnari.oz.au (Pacific Rim), ds.internic.net (US East Coast), or
25	   ftp.isi.edu (US West Coast).

27	   Distribution of this document is unlimited. Please send comments to
28	   the HTTP working group at . Discussions
29	   of the working group are archived at
30	   . General discussions
31	   about HTTP and the applications which use HTTP should take place on
32	   the  mailing list.

34	Abstract

36	   The Hypertext Transfer Protocol (HTTP) is an application-level
37	   protocol with the lightness and speed necessary for distributed,
38	   collaborative, hypermedia information systems. It is a generic,
39	   stateless, object-oriented protocol which can be used for many
40	   tasks, such as name servers and distributed object management
41	   systems, through extension of its request methods (commands). A
42	   feature of HTTP is the typing and negotiation of data
43	   representation, allowing systems to be built independently of the
44	   data being transferred.

46	   HTTP has been in use by the World-Wide Web global information
47	   initiative since 1990. This specification reflects preferred usage
48	   of the protocol referred to as "HTTP/1.0", and is compatible with
49	   the most commonly used HTTP server and client programs implemented
50	   prior to November 1994.

52	Table of Contents

54	   1.  Introduction
55	       1.1  Purpose
56	       1.2  Overall Operation
57	       1.3  Terminology
58	   2.  Notational Conventions and Generic Grammar
59	       2.1  Augmented BNF
60	       2.2  Basic Rules
61	   3.  Protocol Parameters
62	       3.1  HTTP Version
63	       3.2  Universal Resource Identifiers
64	       3.3  Date/Time Formats
65	            3.3.1  Full Date
66	            3.3.2  Delta Seconds
67	   4.  HTTP Message
68	       4.1  Message Types
69	       4.2  Message Headers
70	       4.3  General Message Header Fields
71	            4.3.1  Date
72	            4.3.2  Forwarded
73	            4.3.3  Message-ID
74	            4.3.4  MIME-Version
75	   5.  Request
76	       5.1  Request-Line
77	       5.2  Method
78	            5.2.1  GET
79	            5.2.2  HEAD
80	            5.2.3  POST
81	            5.2.4  PUT
82	            5.2.5  DELETE
83	            5.2.6  LINK
84	            5.2.7  UNLINK
85	       5.3  Request-URI
86	       5.4  Request Header Fields
87	            5.4.1  Accept
88	            5.4.2  Accept-Charset
89	            5.4.3  Accept-Encoding
90	            5.4.4  Accept-Language
91	            5.4.5  Authorization
92	            5.4.6  From
93	            5.4.7  If-Modified-Since
94	            5.4.8  Pragma
95	            5.4.9  Referer
96	            5.4.10 User-Agent
97	   6.  Response
98	       6.1  Status-Line
99	       6.2  Status Codes and Reason Phrases
100	            6.2.1  Successful 2xx
101	            6.2.2  Redirection 3xx
102	            6.2.3  Client Error 4xx
103	            6.2.4  Server Errors 5xx
104	       6.3  Response Header Fields
105	            6.3.1  Public
106	            6.3.2  Retry-After
107	            6.3.3  Server
108	            6.3.4  WWW-Authenticate
109	   7.  Entity
110	       7.1  Entity Header Fields
111	            7.1.1  Allow
112	            7.1.2  Content-Encoding
113	            7.1.3  Content-Language
114	            7.1.4  Content-Length
115	            7.1.5  Content-Transfer-Encoding
116	            7.1.6  Content-Type
117	            7.1.7  Derived-From
118	            7.1.8  Expires
119	            7.1.9  Last-Modified
120	            7.1.10 Link
121	            7.1.11 Location
122	            7.1.12 Title
123	            7.1.13 URI
124	            7.1.14 Version
125	       7.2  Entity Body
126	            7.2.1  Type
127	            7.2.2  Length
128	   8.  Content Parameters
129	       8.1  Media Types
130	            8.1.1  Canonicalization and Text Defaults
131	            8.1.2  Multipart Types
132	       8.2  Language Tags
133	       8.3  Character Sets
134	       8.4  Encoding Mechanisms
135	       8.5  Transfer Encodings
136	   9.  Content Negotiation
137	   10. Access Authentication
138	       10.1 Basic Authentication Scheme
139	   11. Security Considerations
140	       11.1 Authentication of Clients
141	       11.2 Idempotent Methods
142	       11.3 Abuse of Server Log Information
143	   12. Acknowledgments
144	   13. References
145	   14. Authors' Addresses
146	   Appendix A.   Internet Media Type message/http
147	   Appendix B.   Minimum Compliance
148	       B.1  Servers
149	   Appendix C.   Tolerant Applications
150	       C.1  Request-Line, Status-Line, and Header Fields
151	   Appendix D.   Relationship to MIME
152	       D.1  Conversion to Canonical Form
153	            D.1.1  Representation of Line Breaks
154	            D.1.2  Default Character Set
155	       D.2  Default Content-Transfer-Encoding
156	       D.3  Introduction of Content-Encoding
157	   Appendix E.   Example of Version Control

159	1.  Introduction

161	1.1  Purpose

163	   The Hypertext Transfer Protocol (HTTP) is an application-level
164	   protocol with the lightness and speed necessary for distributed,
165	   collaborative, hypermedia information systems. HTTP has been in use
166	   by the World-Wide Web global information initiative since 1990.
167	   This specification reflects preferred usage of the protocol
168	   referred to as "HTTP/1.0". This specification does not necessarily
169	   reflect the "current practice" of any single HTTP server or client
170	   implementation. It does, however, seek to remain compatible with
171	   existing implementations wherever possible, and should be
172	   considered the reference for future implementations of HTTP/1.0.

174	   Practical information systems require more functionality than
175	   simple retrieval, including search, front-end update, and
176	   annotation. HTTP/1.0 allows an open-ended set of methods to be used
177	   to indicate the purpose of a request. It builds on the discipline
178	   of reference provided by the Universal Resource Identifier
179	   (URI) [3], as a location (URL) [5] or name (URN), for indicating
180	   the resource on which a method is to be applied. Messages are
181	   passed in a format similar to that used by Internet Mail [8] and
182	   the Multipurpose Internet Mail Extensions (MIME) [6].

184	   HTTP/1.0 is also used for communication between user agents and
185	   various gateways, allowing hypermedia access to existing Internet
186	   protocols like SMTP [14], NNTP [12], FTP [16], Gopher [2], and
187	   WAIS [9]. HTTP/1.0 is designed to allow such gateways, via proxy
188	   servers, without any loss of the data conveyed by those earlier
189	   protocols.

191	1.2  Overall Operation

193	   The HTTP protocol is based on a request/response paradigm. A
194	   requesting program (termed a client) establishes a connection with
195	   a receiving program (termed a server) and sends a request to the
196	   server in the form of a request method, URI, and protocol version,
197	   followed by a MIME-like message containing request modifiers,
198	   client information, and possible body content. The server responds
199	   with a status line (including its protocol version and a success or
200	   error code), followed by a MIME-like message containing server
201	   information, entity metainformation, and possible body content. It
202	   should be noted that a given program may be capable of being both a
203	   client and a server; our use of those terms refers only to the role
204	   being performed by the program during a particular connection,
205	   rather than to the program's purpose in general.

207	   On the Internet, the communication generally takes place over a
208	   TCP/IP connection. The default port is TCP 80 [17], but other ports
209	   can be used. This does not preclude the HTTP/1.0 protocol from
210	   being implemented on top of any other protocol on the Internet, or
211	   on other networks. The mapping of the HTTP/1.0 request and response
212	   structures onto the transport data units of the protocol in
213	   question is outside the scope of this specification.

215	   For most implementations, the connection is established by the
216	   client prior to each request and closed by the server after sending
217	   the response. However, this is not a feature of the protocol and is
218	   not required by this specification. Both clients and servers must
219	   be capable of handling cases where either party closes the
220	   connection prematurely, due to user action, automated time-out, or
221	   program failure. In any case, the closing of the connection by
222	   either or both parties always terminates the current request,
223	   regardless of its status.

225	1.3  Terminology

227	   This specification uses a number of terms to refer to the roles
228	   played by participants in, and objects of, the HTTP communication.

230	   connection
231	       A virtual circuit established between two parties for the
232	       purpose of communication.

234	   message
235	       A structured sequence of octets transmitted via the connection
236	       as the basic component of communication.

238	   request
239	       An HTTP request message (as defined in Section 5).

241	   response
242	       An HTTP response message (as defined in Section 6).

244	   resource
245	       A network data object or service which can be identified by a
246	       URI.

248	   entity
249	       A particular representation or rendition of a resource that may
250	       be enclosed within a request or response message. An entity
251	       consists of metainformation (in the form of entity headers) and
252	       content (in the form of an entity body).

254	   client
255	       A program that establishes connections for the purpose of
256	       sending requests.

258	   user agent
259	       The client program which is closest to the user and which
260	       initiates requests at their behest.

262	   server
263	       A program that accepts connections in order to service requests
264	       by sending back responses.

266	   origin server
267	       The server on which a given resource resides or is to be created.

269	   proxy
270	       An intermediary program which acts as both a server and a client
271	       for the purpose of forwarding requests. Proxies are often used
272	       to act as a portal through a network firewall. A proxy server
273	       accepts requests from other clients and services them either
274	       internally or by passing them (with possible translation) on to
275	       other servers. A caching proxy is a proxy server with a local
276	       cache of server responses -- some requested resources can be
277	       serviced from the cache rather than from the origin server. Some
278	       proxy servers also act as origin servers.

280	   gateway
281	       A proxy which services HTTP requests by translation into
282	       protocols other than HTTP. The reply sent from the remote server
283	       to the gateway is likewise translated into HTTP before being
284	       forwarded to the user agent.

286	2.  Notational Conventions and Generic Grammar

288	2.1  Augmented BNF

290	   All of the mechanisms specified in this document are described in
291	   both prose and an augmented Backus-Naur Form (BNF) similar to that
292	   used by RFC 822 [8]. Implementors will need to be familiar with the
293	   notation in order to understand this specification. The augmented
294	   BNF includes the following constructs:

296	   name = definition

298	       The name of a rule is simply the name itself (without any
299	       enclosing "<" and ">") and is separated from its definition by
300	       the equal character "=". Whitespace is only significant in that
301	       indentation of continuation lines is used to indicate a rule
302	       definition that spans more than one line. Certain basic rules
303	       are in uppercase, such as SP, LWS, HTAB, CRLF, DIGIT, ALPHA,
304	       etc. Angle brackets are used within definitions whenever their
305	       presence will facilitate discerning the use of rule names.

307	   "literal"

309	       Quotation marks surround literal text. Unless stated otherwise,
310	       the text is case-insensitive.

312	   rule1 | rule2

314	       Elements separated by a bar ("I") are alternatives,
315	       e.g. "yes | no" will accept yes or no.

317	   (rule1 rule2)

319	       Elements enclosed in parentheses are treated as a single
320	       element. Thus,"(elem (foo | bar) elem)" allows the token
321	       sequences "elem foo elem" and "elem bar elem".

323	   *rule

325	       The character "*" preceding an element indicates repetition. The
326	       full form is "*element" indicating at least  and at
327	       most  occurrences of element. Default values are 0 and
328	       infinity so that "*(element)" allows any number, including zero;
329	       "1*element" requires at least one; and "1*2element" allows one
330	       or two.

332	   [rule]

334	       Square brackets enclose optional elements; "[foo bar]" is
335	       equivalent to "*1(foo bar)".

337	   N rule

339	       Specific repetition: "(element)" is equivalent to
340	       "*(element)"; that is, exactly  occurrences of
341	       (element). Thus 2DIGIT is a 2-digit number, and 3ALPHA is a
342	       string of three alphabetic characters.

344	   #rule

346	       A construct "#" is defined, similar to "*", for defining lists
347	       of elements. The full form is "#element" indicating at
348	       least  and at most  elements, each separated by one or
349	       more commas (",") and optional linear whitespace (LWS). This
350	       makes the usual form of lists very easy; a rule such as
351	       "( *LWS element *( *LWS "," *LWS element ))" can be shown as
352	       "1#element". Wherever this construct is used, null elements are
353	       allowed, but do not contribute to the count of elements present.
354	       That is, "(element), , (element)" is permitted, but counts as
355	       only two elements. Therefore, where at least one element is
356	       required, at least one non-null element must be present. Default
357	       values are 0 and infinity so that "#(element)" allows any
358	       number, including zero; "1#element" requires at least one; and
359	       "1#2element" allows one or two.

361	   ; comment

363	       A semi-colon, set off some distance to the right of rule text,
364	       starts a comment that continues to the end of line. This is a
365	       simple way of including useful notes in parallel with the
366	       specifications.

368	   implied *LWS

370	       The grammar described by this specification is word-based.
371	       Except where noted otherwise, zero or more linear whitespace
372	       (LWS) can be included between any two words (token or
373	       quoted-string) without changing the interpretation of a field.
374	       However, applications should attempt to follow "common form"
375	       when generating HTTP constructs, since there exist some
376	       implementations that fail to accept anything beyond the common
377	       forms.

379	2.2  Basic Rules

381	   The following rules are used throughout this specification to
382	   describe basic parsing constructs. The US-ASCII character set is
383	   defined by [18].

385	       OCTET          = 
386	       CHAR           = 
387	       UPALPHA        = 
388	       LOALPHA        = 
389	       ALPHA          = UPALPHA | LOALPHA
390	       DIGIT          = 
391	       CTL            = 
393	       CR             = 
394	       LF             = 
395	       SP             = 
396	       HTAB           = 
397	       <">            = 

399	   HTTP/1.0 defines the octet sequence CR LF as the end-of-line marker
400	   for all protocol elements except the Entity-Body (see Appendix C
401	   for tolerant applications). The end-of-line marker for an
402	   Entity-Body is defined by its associated media type, as described
403	   in Section 8.1.

405	       CRLF           = CR LF

407	   HTTP/1.0 headers can be folded onto multiple lines if the
408	   continuation lines begin with linear whitespace characters. All
409	   linear whitespace (including folding) has the same semantics as SP.

411	       LWS            = [CRLF] 1*( SP | HTAB )

413	   Many HTTP/1.0 header field values consist of words separated by LWS
414	   or special characters. These special characters must be in a quoted
415	   string to be used within a parameter value.

417	       word           = token | quoted-string

419	       token          = 1*

421	       tspecials      = "(" | ")" | "<" | ">" | "@"
422	                      | "," | ";" | ":" | "\" | <">
423	                      | "/" | "[" | "]" | "?" | "="
424	                      | SP | HTAB

426	   A string of text is parsed as a single word if it is quoted using
427	   double-quote marks or angle brackets.

429	       quoted-string  = ( <"> *(qdtext) <"> )
430	                      | ( "<" *(qatext) ">" )

432	       qdtext         =  and CTLs,
433	                        but including LWS>

435	       qatext         = ", and CTLs,
436	                        but including LWS>

438	   The text rule is only used for descriptive field contents. Words of
439	   *text may contain characters from character sets other than
440	   US-ASCII only when encoded according to the rules of RFC 1522 [13].

442	       text           = 

445	3.  Protocol Parameters

447	3.1  HTTP Version

449	   HTTP uses a "." numbering scheme to indicate versions
450	   of the protocol. The protocol versioning policy is intended to
451	   allow the sender to indicate the format of a message and its
452	   capacity for understanding further HTTP communication, rather than
453	   the features obtained via that communication. No change is made to
454	   the version number for the addition of message components which do
455	   not affect communication behavior or which only add to extensible
456	   field values. The  number is incremented when the changes
457	   made to the protocol add features which do not change the general
458	   message parsing algorithm, but which may add to the message
459	   semantics and imply additional capabilities of the sender. The
460	    number is incremented when the format of a message within
461	   the protocol is changed.

463	   The version of an HTTP message is indicated by an HTTP-Version
464	   field in the first line of the message. If the protocol version is
465	   not specified, the recipient must assume that the message is in the
466	   simple HTTP/0.9 format.

468	       HTTP-Version   = "HTTP" "/" 1*DIGIT "." 1*DIGIT

470	   Note that the major and minor numbers should be treated as separate
471	   integers and that each may be incremented higher than a single
472	   digit. Thus, HTTP/2.4 is a lower version than HTTP/2.13, which in
473	   turn is lower than HTTP/12.3. Leading zeroes should be ignored by
474	   recipients and never generated by senders.

476	   This document defines both the 0.9 and 1.0 versions of the HTTP
477	   protocol. Applications sending Full-Request or Full-Response
478	   messages, as defined by this specification, must include an
479	   HTTP-Version of "HTTP/1.0".

481	   HTTP servers are required to be able to recognize the format of the
482	   Request-Line for all lower-version requests, understand requests
483	   with a format within one major number of their native version (i.e.
484	    and ), and respond appropriately with a message
485	   within the same  protocol number (even if the response is
486	   simply an error message). HTTP clients are required to be able to
487	   recognize the format of the Status-Line for all lower-version
488	   responses and understand responses with a format within one major
489	   number of their request version. The following hypothetical example
490	   illustrates the required behavior.

492	      o A valid HTTP/3.5 request is received and the server's native
493	        protocol version is

495	          o Less than 3.0: it should attempt to understand the request
496	            and respond (possibly with an error) in its native format;

498	          o Major number of 3: It should understand the request and
499	            respond in its native format;

501	          o Major number of 4: It should understand the request and
502	            respond with a version 3 message;

504	          o Major number higher than 4: It should attempt to understand
505	            the request and respond (possibly with an error) with a
506	            version 3 message;

508	      o User agent receives a response to an HTTP/3.5 request, and the
509	        response version is

511	          o Less than 2.0: It should attempt to understand the response
512	            and unobtrusively warn the user of the version mismatch;

514	          o 2.0--3.4: It should understand the response and be aware
515	            that its request may not have been fully understood by the
516	            server;

518	          o 3.5 or higher 3: It should understand the response and can
519	            assume that the server understood all aspects of the request
520	            if the response does not indicate an error;

522	          o 4.0 or higher: It should attempt to understand the response
523	            and unobtrusively warn the user of the version mismatch.

525	   Proxies must be careful in forwarding requests that are received in
526	   a format different than that of the proxy's native version. Since
527	   the protocol version indicates the protocol capability of the
528	   sender, a proxy must never send a message with a version indicator
529	   which is greater than its native version; if a higher version
530	   request is received, the proxy must either downgrade the request
531	   version or respond with an error. Requests with a version lower
532	   than that of the proxy's native format may be upgraded by the proxy
533	   before being forwarded; the proxy's response to that request must
534	   follow the normal server requirements.

536	3.2  Universal Resource Identifiers

538	   This section is To-Be-Specified. It will contain a brief
539	   description of URIs as defined by RFC 1630, including mention of
540	   the allowed characters and the hex encoding, and provide a full
541	   definition of the http URL scheme.

543	       URI            = 

545	   All URI protocol elements must be encoded using the escaping scheme
546	   described in RFC 1630 [3].

548	3.3  Date/Time Formats

550	3.3.1 Full Date

552	   HTTP/1.0 applications have historically allowed three different
553	   formats for the representation of date/time stamps:

555	       Sun, 06 Nov 1994 08:49:37 GMT   ; RFC 822, updated by RFC 1123
556	       Sunday, 06-Nov-94 08:49:37 GMT  ; RFC 850, obsoleted by RFC 1036
557	       Sun Nov  6 08:49:37 1994        ; ANSI C's asctime() format

559	   The first format is preferred as an Internet standard and
560	   represents a fixed-length subset of that defined by RFC 1123 [7]
561	   (an update to RFC 822 [8]). The second format is in common use, but
562	   is based on the obsolete RFC 850 [11] date format and lacks a four-
563	   digit year. HTTP/1.0 clients and servers must accept all three
564	   formats, though they should never generate the third (asctime)
565	   format. Future clients and servers must only generate the RFC 1123
566	   format for representing date/time stamps in HTTP/1.0 requests and
567	   responses.

569	   All HTTP/1.0 date/time stamps must be represented in Universal Time
570	   (UT), also known as Greenwich Mean Time (GMT), without exception.
571	   This is indicated in the first two formats by the inclusion of
572	   "GMT" as the three-letter abbreviation for time zone, and should be
573	   assumed when reading the asctime format.

575	       HTTP-date      = rfc1123-date | rfc850-date | asctime-date

577	       rfc1123-date   = wkday "," SP date1 SP time SP "GMT"
578	       rfc850-date    = weekday "," SP date2 SP time SP "GMT"
579	       asctime-date   = wkday SP date3 SP time SP 4DIGIT

581	       date1          = 2DIGIT SP month SP 4DIGIT
582	                        ; day month year (e.g. 02 Jun 1982)
583	       date2          = 2DIGIT "-" month "-" 2DIGIT
584	                        ; day-month-year (e.g. 02-Jun-82)
585	       date3          = month SP ( 2DIGIT | ( SP 1DIGIT ))
586	                        ; month day (e.g. Jun  2)

588	       time           = 2DIGIT ":" 2DIGIT ":" 2DIGIT
589	                        ; 00:00:00 - 23:59:59

591	       wkday          = "Mon" | "Tue" | "Wed"
592	                      | "Thu" | "Fri" | "Sat" | "Sun"

594	       weekday        = "Monday" | "Tuesday" | "Wednesday"
595	                      | "Thursday" | "Friday" | "Saturday" | "Sunday"

597	       month          = "Jan" | "Feb" | "Mar" | "Apr"
598	                      | "May" | "Jun" | "Jul" | "Aug"
599	                      | "Sep" | "Oct" | "Nov" | "Dec"

601	   Comments and/or extra LWS are not permitted inside an HTTP-date
602	   value generated by a conformant application. However, recipients of
603	   date values should be robust in accepting date values that may have
604	   been generated by non-HTTP applications (as is sometimes the case
605	   when retrieving or posting messages via gateways to SMTP or NNTP).

607	       Note: HTTP/1.0 requirements for the date/time stamp format
608	       apply only to their usage within the protocol stream.
609	       Clients and servers are not required to use these formats
610	       for user presentation, request logging, etc.

612	3.3.2 Delta Seconds

614	   Some HTTP header fields allow a time value to be specified as an
615	   integer number of seconds (represented in decimal) after the time
616	   that the message was received. This format should only be used to
617	   represent short time periods or periods that cannot start until
618	   receipt of the message.

620	       delta-seconds  = 1*DIGIT

622	4.  HTTP Message

624	4.1  Message Types

626	   HTTP messages consist of requests from client to server and
627	   responses from server to client.

629	       HTTP-message   = Simple-Request           ; HTTP/0.9 messages
630	                      | Simple-Response
631	                      | Full-Request             ; HTTP/1.0 messages
632	                      | Full-Response

634	   Full-Request and Full-Response use the generic message format of
635	   RFC 822 [8] for transferring entities. Both messages may include
636	   optional header fields (a.k.a. "headers") and an entity body. The
637	   entity body is separated from the headers by a null line (i.e., a
638	   line with nothing preceding the CRLF).

640	       Full-Request   = Request-Line              ; Section 5.1
641	                        *( General-Header         ; Section 4.3
642	                        |  Request-Header         ; Section 5.4
643	                        |  Entity-Header )        ; Section 7.1
644	                        CRLF
645	                        [ Entity-Body ]           ; Section 7.2

647	       Full-Response  = Status-Line               ; Section 6.1
648	                        *( General-Header         ; Section 4.3
649	                        |  Response-Header        ; Section 6.3
650	                        |  Entity-Header )        ; Section 7.1
651	                        CRLF
652	                        [ Entity-Body ]           ; Section 7.2

654	   Simple-Request and Simple-Response do not allow the use of any
655	   header information and are limited to a single request method (GET).

657	       Simple-Request  = "GET" SP Request-URI CRLF

659	       Simple-Response = [ Entity-Body ]

661	   Use of the Simple-Request format is discouraged because it prevents
662	   the client from using content negotiation and the server from
663	   identifying the media type of the returned entity.

665	4.2  Message Headers

667	   HTTP header fields, which include General-Header (Section 4.3),
668	   Request-Header (Section 5.4), Response-Header (Section 6.3), and
669	   Entity-Header (Section 7.1) fields, follow the same generic format
670	   as that given in Section 3.1 of RFC 822 [8]. Each header field
671	   consists of a name followed by a colon (":") and the field value.
672	   The field value may be preceded by any amount of LWS, though a
673	   single SP is preferred. Header fields can be extended over multiple
674	   lines by preceding each extra line with at least one LWS.

676	       HTTP-header    = field-name ":" [ field-value ] CRLF

678	       field-name     = 1*
679	       field-value    = *( field-content | comment | LWS )

681	       field-content  = 

685	   The order in which header fields are received is not significant.
686	   However, it is considered "good practice" to send General-Header
687	   fields first, followed by Request-Header or Response-Header fields
688	   prior to the Entity-Header fields. Comments can be included in HTTP
689	   header fields by surrounding the comment text with parentheses.

691	       comment        = "(" *( ctext | comment ) ")"
692	       ctext          = 

694	       Note: Use of comments within HTTP headers is generally
695	       discouraged, since they are rarely seen by human eyes and
696	       hence only increase network traffic. However, they may be
697	       useful for messages posted or retrieved via NNTP and SMTP
698	       gateways.

700	4.3  General Message Header Fields

702	   There are a few header fields which have general applicability for
703	   both request and response messages, but which do not apply to the
704	   communicating parties or the entity being transferred. Although
705	   none of the General-Header fields are required, they are all
706	   strongly recommended where their use is appropriate, and should be
707	   understood by all future HTTP/1.0 clients and servers. These
708	   headers apply only to the message being transmitted.

710	       General-Header = Date
711	                      | Forwarded
712	                      | Message-ID
713	                      | MIME-Version
714	                      | extension-header

716	   Although additional general header fields may be implemented via
717	   the extension mechanism, applications which do not recognize those
718	   fields should treat them as Entity-Header fields.

720	4.3.1 Date

722	   The Date header represents the date and time at which the message
723	   was originated, having the same semantics as orig-date in RFC
724	   822. The field value is an HTTP-date, as described in Section 3.3.

726	       Date           = "Date" ":" HTTP-date

728	   An example is

730	       Date: Tue, 15 Nov 1994 08:12:31 GMT

732	   If a message is received via direct connection with the user agent
733	   (in the case of requests) or the origin server (in the case of
734	   responses), then the default date can be assumed to be the current
735	   date at the receiving end. However, since the date--as it is
736	   believed by the origin--is important for evaluating cached
737	   responses, origin servers should always include a Date header.
738	   Clients should only send a Date header field in messages that
739	   include an entity body, as in the case of the PUT and POST
740	   requests, and even then it is optional. A received message which
741	   does not have a Date header field should be assigned one by the
742	   receiver if and only if the message will be cached by that receiver
743	   or gatewayed via a protocol which requires a Date.

745	   Only one Date header field is allowed per message. In theory, the
746	   date should represent the moment just before the status/request-
747	   line is generated (i.e. the time at which the originator made the
748	   determination of what the request/response should be). In practice,
749	   the date can be generated at any time during the message
750	   origination without affecting its semantic value.

752	       Note: An earlier version of this document incorrectly
753	       specified that this field should contain the creation date
754	       of the enclosed Entity-Body. This has been changed to
755	       reflect actual (and proper) usage.

757	4.3.2 Forwarded

759	   The Forwarded header is to be used by proxies to indicate the
760	   intermediate steps between the user agent and the server (on
761	   requests) and between the origin server and the client (on
762	   responses). It is analogous to the "Received" field of RFC 822 and
763	   is intended to be used for tracing transport problems and avoiding
764	   request loops.

766	       Forwarded      = "Forwarded" ":" "by" URI [ "(" product ")" ]
767	                        [ "for" FQDN ]

769	       FQDN           = 

771	   For example, a message could be sent from a client on
772	   ptsun00.cern.ch to a server at www.ics.uci.edu port 80, via an
773	   intermediate HTTP proxy at info.cern.ch port 8000. The request
774	   received by the server at www.ics.uci.edu would then have the
775	   following Forwarded header field:

777	       Forwarded: by http://info.cern.ch:8000/ for ptsun00.cern.ch

779	   Multiple Forwarded header fields are allowed and should represent
780	   each proxy that has forwarded the message. It is strongly
781	   recommended that proxies used as a portal through a network
782	   firewall do not, by default, send out information about the
783	   internal hosts within the firewall region. This information should
784	   only be propagated if explicitly enabled. If not enabled, the for
785	   token and FQDN should not be included in the field value.

787	4.3.3 Message-ID

789	   The Message-ID field in HTTP is identical to that used by Internet
790	   Mail and USENET messages, as defined in [11]. That is, it gives the
791	   message a single, unique identifier which can be used for
792	   identifying the message (not its contents) for "much longer" than
793	   the expected lifetime of that message.

795	       Message-ID     = "Message-ID" ":" "<" addr-spec ">"
796	       addr-spec      = 

798	   Although it is not required, the addr-spec format typically used
799	   within a Message-ID consists of a string that is unique at the
800	   originator's machine, followed by the required "@" character and
801	   the fully-qualified domain name of that machine. An example is

803	       Message-ID: <9411151630.4256@info.cern.ch>

805	   which is composed using the time, date and process-ID on the host
806	   info.cern.ch. However, this is only one of many possible methods
807	   for generating a unique Message-ID and recipients of a message
808	   should consider the entire value opaque.

810	   The Message-ID field is normally not generated by HTTP applications
811	   and is never required. It should only be generated by a gateway
812	   application when the message is being posted to some other protocol
813	   that desires a Message-ID. HTTP responses should only include a
814	   Message-ID header field when the entity being transferred already
815	   has one assigned to it (as in the case of resources that were
816	   originally posted via Internet Mail or USENET).

818	4.3.4 MIME-Version

820	   HTTP is not a MIME-conformant protocol. However, HTTP/1.0 messages
821	   may include a single MIME-Version header field to indicate what
822	   version of the MIME protocol was used to construct the message. Use
823	   of the MIME-Version header field should indicate that the message
824	   is in full compliance with the MIME protocol (as defined in [6]).
825	   Unfortunately, current versions of HTTP/1.0 clients and servers use
826	   this field indiscriminately, and thus receivers must not take it
827	   for granted that the message is indeed in full compliance with
828	   MIME. Gateways are responsible for ensuring this compliance (where
829	   possible) when exporting HTTP messages to strict MIME environments.
830	   Future HTTP/1.0 applications must only use MIME-Version when the
831	   message is intended to be MIME-conformant.

833	       MIME-Version   = "MIME-Version" ":" 1*DIGIT "." 1*DIGIT

835	   MIME version "1.0" is the default for use in HTTP/1.0. However,
836	   HTTP/1.0 message parsing and semantics are defined by this document
837	   and not the MIME specification.

839	5. Request

841	   A request message from a client to a server includes, within the
842	   first line of that message, the method to be applied to the
843	   resource requested, the identifier of the resource, and the
844	   protocol version in use. For backwards compatibility with the more
845	   limited HTTP/0.9 protocol, there are two valid formats for an HTTP
846	   request:

848	       Request        = Simple-Request | Full-Request

850	       Simple-Request = "GET" SP Request-URI CRLF

852	       Full-Request   = Request-Line              ; Section 5.1
853	                        *( General-Header         ; Section 4.3
854	                        |  Request-Header         ; Section 5.4
855	                        |  Entity-Header )        ; Section 7.1
856	                        CRLF
857	                        [ Entity-Body ]           ; Section 7.2

859	   If an HTTP/1.0 server receives a Simple-Request, it must respond
860	   with an HTTP/0.9 Simple-Response. An HTTP/1.0 client capable of
861	   receiving a Full-Response should never generate a Simple-Request.

863	5.1  Request-Line

865	   The Request-Line begins with a method token, followed by the
866	   Request-URI and the protocol version, and ending with CRLF. The
867	   elements are separated by SP characters. No CR or LF are allowed
868	   except in the final CRLF sequence.

870	       Request-Line   = Method SP Request-URI SP HTTP-Version CRLF

872	   Note that the difference between a Simple-Request and the Request-
873	   Line of a Full-Request is the presence of the HTTP-Version field.

875	5.2  Method

877	   The Method token indicates the method to be performed on the
878	   resource identified by the Request-URI. The method is case-
879	   sensitive and extensible.

881	       Method         = "GET" | "HEAD" | "PUT" | "POST"
882	                      | "DELETE" | "LINK" | "UNLINK"
883	                      | extension-method

885	       extension-method = token

887	   The list of methods acceptable by a specific resource can be
888	   specified in an "Allow" Entity-Header (Section 7.1.1). However, the
889	   client is always notified through the return code of the response
890	   whether a method is currently allowed on a specific resource, as
891	   this can change dynamically. Servers should return the status code
892	   "405 Method Not Allowed" if the method is known by the server but
893	   not allowed for the requested resource, and "501 Not Implemented"
894	   if the method is unknown or not implemented by the server.

896	   The set of common methods for HTTP/1.0 is described below. Although
897	   this set can be easily expanded, additional methods cannot be
898	   assumed to share the same semantics for separately extended clients
899	   and servers. In order to maintain compatibility, the semantic
900	   definition for extension methods should be registered with the
901	   IANA [17].

903	5.2.1 GET

905	   The GET method means retrieve whatever information (in the form of
906	   an entity) is identified by the Request-URI. If the Request-URI
907	   refers to a data-producing process, it is the produced data which
908	   shall be returned as the entity in the response and not the source
909	   text of the process (unless that text happens to be the output of
910	   the process).

912	   The semantics of the GET method changes to a "conditional GET" if
913	   the request message includes an If-Modified-Since header field. A
914	   conditional GET method requests that the identified resource be
915	   transferred only if it has been modified since the date given by
916	   the If-Modified-Since header. The algorithm for determining this
917	   includes the following cases:

919	      a)   If the request would normally result in anything other than
920	           a "200 OK" status, or if the passed If-Modified-Since date
921	           is invalid, the response is exactly the same as for a
922	           normal GET.

924	      b)   If the resource has been modified since the If-Modified-
925	           Since date, the response is exactly the same as for a
926	           normal GET.

928	      c)   If the resource has not been modified since the If-Modified-
929	           Since date, the server shall return a "304 Not Modified"
930	           response.

932	   The conditional GET method is intended to reduce network usage by
933	   allowing cached entities to be refreshed without requiring multiple
934	   requests or transferring unnecessary data.

936	5.2.2 HEAD

938	   The HEAD method is identical to GET except that the server must not
939	   return any Entity-Body in the response. The metainformation
940	   contained in the HTTP headers in response to a HEAD request should
941	   be identical to the information sent in response to a GET request.
942	   This method can be used for obtaining metainformation about the
943	   resource identified by the Request-URI without transferring the
944	   Entity-Body itself. This method is often used for testing hypertext
945	   links for validity, accessibility, and recent modification.

947	   There is no "conditional HEAD" requests analogous to the
948	   conditional GET. If an If-Modified-Since header field is included
949	   with a HEAD request, it should be ignored.

951	5.2.3 POST

953	   The POST method is used to request that the destination server
954	   accept the entity enclosed in the request as a new subordinate of
955	   the resource identified by the Request-URI in the Request-Line.
956	   POST is designed to allow a uniform method to cover the following
957	   functions:

959	      o Annotation of existing resources;

961	      o Posting a message to a bulletin board, newsgroup, mailing list,
962	        or similar group of articles;

964	      o Providing a block of data (usually a form) to a data-handling
965	        process;

967	      o Extending a database through an append operation.

969	   The actual function performed by the POST method is determined by
970	   the server and is usually dependent on the Request-URI. The posted
971	   entity is considered to be subordinate to that URI in the same way
972	   that a file is subordinate to a directory containing it, a news
973	   article is subordinate to a newsgroup to which it is posted, or a
974	   record is subordinate to a database.

976	   The client can suggest a URI for identifying the new resource by
977	   including a URI-header field in the request. However, the server
978	   should treat that URI as advisory only and may store the entity
979	   under a different URI or without any URI.

981	   The client may apply relationships between the new resource and
982	   other existing resources by including Link header fields, as
983	   described in Section 7.1.10. The server may use the Link
984	   information to perform other operations as a result of the new
985	   resource being added. For example, lists and indexes might be
986	   updated. However, no mandatory operation is imposed on the origin
987	   server. The origin server may also generate its own or additional
988	   links to other resources.

990	   A successful POST does not require that the entity be created as a
991	   resource on the origin server or made accessible for future
992	   reference. That is, the action performed by the POST method might
993	   not result in a resource that can be identified by a URI. In this
994	   case, either "200 OK" or "204 No Content" is the appropriate
995	   response status, depending on whether or not the response includes
996	   an entity that describes the result.

998	   If a resource has been created on the origin server, the response
999	   should be "201 Created" and contain the allocated URI, all
1000	   applicable Link header fields, and an entity (preferably of type
1001	   "text/html") which describes the status of the request and refers
1002	   to the new resource.

1004	5.2.4 PUT

1006	   The PUT method requests that the enclosed entity be stored under
1007	   the supplied Request-URI. If the Request-URI refers to an already
1008	   existing resource, the enclosed entity should be considered a
1009	   modified version of the one residing on the origin server. If the
1010	   Request-URI does not point to an existing resource, and that URI is
1011	   capable of being defined as a new resource by the requesting user
1012	   agent, the origin server can create the resource with that URI. If
1013	   a new resource is created, the origin server must inform the user
1014	   agent via the "201 Created" response. If an existing resource is
1015	   modified, the "200 OK" response should be sent to indicate
1016	   successful completion of the request. If the resource could not be
1017	   created or modified with the Request-URI, an appropriate error
1018	   response should be given that reflects the nature of the problem.

1020	   The fundamental difference between the POST and PUT requests is
1021	   reflected in the different meaning of the Request-URI. The URI in a
1022	   POST request identifies the resource that will handle the enclosed
1023	   entity as an appendage. That resource may be a data-accepting
1024	   process, a gateway to some other protocol, or a separate entity
1025	   that accepts annotations. In contrast, the URI in a PUT request
1026	   identifies the entity enclosed with the request. The requestor of a
1027	   PUT knows what URI is intended and the receiver must not attempt to
1028	   apply the request to some other resource. If the receiver desires
1029	   that the request be applied to a different URI, it must send a
1030	   "301 Moved Permanently" response; the requestor may then make its
1031	   own decision regarding whether or not to redirect the request.

1033	   The client can create or modify relationships between the enclosed
1034	   entity and other existing resources by including Link header
1035	   fields, as described in Section 7.1.10. As with POST, the server
1036	   may use the Link information to perform other operations as a
1037	   result of the request. However, no mandatory operation is imposed
1038	   on the origin server. The origin server may generate its own or
1039	   additional links to other resources.

1041	   The actual method for determining how the resource is placed, and
1042	   what happens to its predecessor, is defined entirely by the origin
1043	   server. If version control is implemented by the origin server, the
1044	   Version and Derived-From header fields should be used to help
1045	   identify and control revisions to a resource.

1047	5.2.5 DELETE

1049	   The DELETE method requests that the origin server delete the
1050	   resource identified by the Request-URI. This method may be
1051	   overridden by human intervention (or other means) on the origin
1052	   server. The client cannot be guaranteed that the operation has been
1053	   carried out, even if the status code returned from the origin
1054	   server indicates that the action has been completed successfully.
1055	   However, the server should not indicate success unless, at the time
1056	   the response is given, it intends to delete the resource or move it
1057	   to an inaccessible location.

1059	   A successful response should be "200 OK" (if the response includes
1060	   an entity describing the status), "202 Accepted" (if the action has
1061	   not yet been enacted), or "204 No Content" (if the response is OK
1062	   but does not include an entity).

1064	5.2.6 LINK

1066	   The LINK method establishes one or more Link relationships between
1067	   the existing resource identified by the Request-URI and other
1068	   existing resources. The difference between LINK and other methods
1069	   allowing links to be established between resources is that the LINK
1070	   method does not allow any Entity-Body to be sent in the request and
1071	   does not result in the creation of new resources.

1073	5.2.7 UNLINK

1075	   The UNLINK method removes one or more Link relationships from the
1076	   existing resource identified by the Request-URI. These
1077	   relationships may have been established using the LINK method or by
1078	   any other method supporting the Link header. The removal of a link
1079	   to a resource does not imply that the resource ceases to exist or
1080	   becomes inaccessible for future references.

1082	5.3  Request-URI

1084	   The Request-URI is a Universal Resource Identifier (Section 3.2)
1085	   and identifies the resource upon which to apply the request.

1087	       Request-URI    = URI

1089	   Unless the server is being used as a proxy, a partial URI shall be
1090	   given with the assumptions of the scheme (http) and hostname:port
1091	   (the server's address) being obvious. That is, if the full URI
1092	   looks like

1094	       http://info.cern.ch/hypertext/WWW/TheProject.html

1096	   then the corresponding partial URI in the Simple-Request or
1097	   Full-Request is

1099	       /hypertext/WWW/TheProject.html

1101	   If the client is sending the request through a proxy, the absolute
1102	   form of the URI (including scheme and server address) must be used.
1103	   A proxy must be able to recognize all of its server names,
1104	   including any aliases, local variations, and the numeric IP address.

1106	5.4  Request Header Fields

1108	   The request header fields allow the client to pass additional
1109	   information about the request (and about the client itself) to the
1110	   server. All header fields are optional and conform to the generic
1111	   HTTP-header syntax.

1113	       Request-Header = Accept
1114	                      | Accept-Charset
1115	                      | Accept-Encoding
1116	                      | Accept-Language
1117	                      | Authorization
1118	                      | From
1119	                      | If-Modified-Since
1120	                      | Pragma
1121	                      | Referer
1122	                      | User-Agent
1123	                      | extension-header

1125	   Although additional request header fields may be implemented via
1126	   the extension mechanism, applications which do not recognize those
1127	   fields should treat them as Entity-Header fields.

1129	5.4.1 Accept

1131	   The Accept header field can be used to indicate a list of media
1132	   ranges which are acceptable as a response to the request. The
1133	   asterisk "*" character is used to group media types into ranges,
1134	   with "*/*" indicating all media types and "type/*" indicating all
1135	   subtypes of that type. The set of ranges given by the client should
1136	   represent what types are acceptable given the context of the
1137	   request. The Accept field should only be used when the request is
1138	   specifically limited to a set of desired types (as in the case of a
1139	   request for an in-line image), to indicate qualitative preferences
1140	   for specific media types, or to indicate acceptance of unusual
1141	   media types.

1143	   The field may be folded onto several lines and more than one
1144	   occurrence of the field is allowed (with the semantics being the
1145	   same as if all the entries had been in one field value).

1147	       Accept         = "Accept" ":" 1#(
1148	                             media-range
1149	                             [ ";" "q" "=" ( "0" | "1" | float ) ]
1150	                             [ ";" "mxb" "=" 1*DIGIT ] )

1152	       media-range    = ( "*/*"
1153	                      |   ( type "/" "*" )
1154	                      |   ( type "/" subtype )
1155	                        ) *( ";" parameter )

1157	       float          = < ANSI-C floating point text representation,
1158	                        where (0.0 < float < 1.0) >

1160	   The parameter q is used to indicate the quality factor, which
1161	   represents the user's preference for that range of media types, and
1162	   the parameter mxb gives the maximum acceptable size of the Entity-
1163	   Body (in decimal number of octets) for that range of media types.
1164	   Section 9 describes the content negotiation algorithm which makes
1165	   use of these values. If at least one Accept header is present, a
1166	   quality factor of 0 is equivalent to not sending an Accept header
1167	   field containing that media-type or set of media-types. The default
1168	   values are: q=1 and mxb=undefined (i.e. infinity).

1170	   The example

1172	       Accept: audio/*; q=0.2, audio/basic

1174	   should verbally be interpreted as "if you have audio/basic, send
1175	   it; otherwise send me any audio type."

1177	   If no Accept header is present, then it is assumed that the client
1178	   accepts all media types with quality factor 1. This is equivalent
1179	   to the client sending the following accept header field:

1181	       Accept: */*; q=1

1183	   or

1185	       Accept: */*

1187	   A more elaborate example is

1189	       Accept: text/plain; q=0.5, text/html,
1190	               text/x-dvi; q=0.8; mxb=100000, text/x-c

1192	   Verbally, this would be interpreted as "text/html and text/x-c are
1193	   the preferred media types, but if they do not exist then send the
1194	   Entity-Body in text/x-dvi if the entity is less than 100000 bytes,
1195	   otherwise send text/plain".

1197	       Note: In earlier versions of this document, the mxs
1198	       parameter defined the maximum acceptable delay in seconds
1199	       before the response would arrive. This has been removed as
1200	       the server has no means of obtaining a useful reference
1201	       value. However, this does not prevent the client from
1202	       internally measuring the response time and optimizing the
1203	       Accept header field accordingly.

1205	   It must be emphasized that this field should only be used when it
1206	   is necessary to restrict the response media types to a subset of
1207	   those possible, to indicate the acceptance of unusual media types,
1208	   or when the user has been permitted to specify qualitative values
1209	   for ranges of media types. If no quality factors have been set by
1210	   the user, and the context of the request is such that the user
1211	   agent is capable of saving the entity to a file if the received
1212	   media type is unknown, then the only appropriate value for Accept
1213	   is "*/*" and the list of unusual types. Whether or not a particular
1214	   media type is deemed "unusual" should be a configurable aspect of
1215	   the user agent.

1217	5.4.2 Accept-Charset

1219	   The Accept-Charset header field can be used to indicate a list of
1220	   preferred character sets other than the default US-ASCII and
1221	   ISO-8859-1. This field allows clients capable of understanding more
1222	   comprehensive or special-purpose character sets to signal that
1223	   capability to a server which is capable of representing documents
1224	   in those character sets.

1226	       Accept-Charset = "Accept-Charset" ":" 1#charset

1228	   Character set values are described in Section 8.3. An example is

1230	       Accept-Charset: iso-8859-5, unicode-1-1

1232	   The value of this field should not include "US-ASCII" or
1233	   "ISO-8859-1", since those values are always assumed by default. If
1234	   a resource is only available in a character set other than the
1235	   defaults, and that character set is not listed in the Accept-
1236	   Charset field, it is only acceptable for the server to send the
1237	   entity if the character set can be identified by an appropriate
1238	   charset parameter on the media type or within the format of the
1239	   media type itself.

1241	       Note: User agents are not required to be able to render the
1242	       characters associated with the ISO-8859-1 character set.
1243	       However, they must be able to interpret their meaning to
1244	       whatever extent is required to properly handle messages in
1245	       that character set..

1247	5.4.3 Accept-Encoding

1249	   The Accept-Encoding header field is similar to Accept, but lists
1250	   the encoding-mechanisms and transfer-encoding values which are
1251	   acceptable in the response.

1253	       Accept-Encoding = "Accept-Encoding" ":"
1254	                         1#( encoding-mechanism | transfer-encoding )

1256	   An example of its use is

1258	       Accept-Encoding: compress, base64, gzip, quoted-printable

1260	   The field value should never include the identity transfer-encoding
1261	   values ("7bit", "8bit", and "binary") since they actually represent
1262	   "no encoding." If no Accept-Encoding field is present in a request,
1263	   it must be assumed that the client does not accept any encoding-
1264	   mechanism and only the identity transfer-encodings.

1266	5.4.4 Accept-Language

1268	   The Accept-Language header field is similar to Accept, but lists
1269	   the set of natural languages that are preferred as a response to
1270	   the request.

1272	       Accept-Language = "Accept-Language" ":" 1#language-tag

1274	   The language-tag is described in Section 8.2. Languages are listed
1275	   in the order of their preference to the user. For example,

1277	       Accept-Language: dk, en-gb

1279	   would mean: "Send me a Danish version if you have it; else a
1280	   British English version."

1282	   If the server cannot fulfill the request with one or more of the
1283	   languages given, or if the languages only represent a subset of a
1284	   multi-linguistic Entity-Body, it is acceptable to serve the request
1285	   in an unspecified language.

1287	       Note: As intelligibility is highly dependent on the
1288	       individual user, it is recommended that client applications
1289	       make the choice of linguistic preference available to the
1290	       user.

1292	5.4.5 Authorization

1294	   A user agent that wishes to authenticate itself with a server
1295	   (usually, but not necessarily, after receiving a "401 Unauthorized"
1296	   response), may do so by including an Authorization header field
1297	   with the request. The Authorization field value consists of
1298	   credentials containing the authentication information of the user
1299	   agent for the realm of the resource being requested.

1301	       Authorization  = "Authorization" ":" credentials

1303	   HTTP access authentication is described in Section 10. If a request
1304	   is authenticated and a realm specified, the same credentials should
1305	   be valid for all other requests within this realm.

1307	5.4.6 From

1309	   The From header field, if given, should contain an Internet e-mail
1310	   address for the human user who controls the requesting user agent.
1311	   The address should be machine-usable, as defined by addr-spec in
1312	   RFC 822:

1314	       From           = "From" ":" addr-spec

1316	   An example is:

1318	       From: webmaster@w3.org

1320	   This header field may be used for logging purposes and as a means
1321	   for identifying the source of invalid or unwanted requests. It
1322	   should not be used as an insecure form of access protection. The
1323	   interpretation of this field is that the request is being performed
1324	   on behalf of the person given, who accepts responsibility for the
1325	   method performed. In particular, robot agents should include this
1326	   header so that the person responsible for running the robot can be
1327	   contacted if problems occur on the receiving end.

1329	   The Internet e-mail address in this field does not have to
1330	   correspond to the Internet host which issued the request. (For
1331	   example, when a request is passed through a proxy, then the
1332	   original issuer's address should be used). The address should, if
1333	   possible, be a valid Internet e-mail address, whether or not it is
1334	   in fact an Internet e-mail address or the Internet e-mail
1335	   representation of an address on some other mail system.

1337	       Note: The client should not send the From header field
1338	       without the user's approval, as it may conflict with the
1339	       user's privacy interests or their site's security policy. It
1340	       is strongly recommended that the user be able to disable,
1341	       enable, and modify the value of this field at any time prior
1342	       to a request.

1344	5.4.7 If-Modified-Since

1346	   The If-Modified-Since header field is used with the GET method to
1347	   make it conditional: if the requested resource has not been
1348	   modified since the time specified in this field, a copy of the
1349	   resource will not be returned from the server; instead, a "304 Not
1350	   Modified" response will be returned without any Entity-Body.

1352	       If-Modified-Since = "If-Modified-Since" ":" HTTP-date

1354	   An example of the field is:

1356	       If-Modified-Since: Sat, 29 Oct 1994 19:43:31 GMT

1358	   The purpose of this feature is to allow efficient updates of local
1359	   cache information with a minimum amount of transaction overhead.
1360	   The same functionality can be obtained, though with much greater
1361	   overhead, by issuing a HEAD request and following it with a GET
1362	   request if the server indicates that the entity has been modified.

1364	5.4.8 Pragma

1366	   The Pragma header field is used to specify directives that must be
1367	   applied to all servers along the request chain (where relevant).
1368	   The directives typically specify behavior that prevents
1369	   intermediate proxies from changing the nature of the request.
1370	   Although multiple pragma directives can be listed as part of the
1371	   request, HTTP/1.0 only defines semantics for the
1372	   "no-cache" directive.

1374	       Pragma           = "Pragma" ":" 1#pragma-directive

1376	       pragma-directive = "no-cache" | extension-pragma
1377	       extension-pragma = token

1379	   When the "no-cache" directive is present, a caching proxy must
1380	   forward the request toward the origin server even if it has a
1381	   cached copy of what is being requested. This allows a client to
1382	   insist upon receiving an authoritative response to its request. It
1383	   also allows a client to refresh a cached copy which has become
1384	   corrupted or is known to be stale.

1386	   Pragmas must be passed through by a proxy even when they have
1387	   significance to that proxy. This is necessary in cases when the
1388	   request has to go through many proxies, and the pragma may affect
1389	   all of them. It is not possible to specify a pragma for a specific
1390	   proxy; however, any pragma-directive not relevant to a gateway or
1391	   proxy should be ignored.

1393	5.4.9 Referer

1395	   The Referer field allows the client to specify, for the server's
1396	   benefit, the address (URI) of the document (or element within the
1397	   document) from which the Request-URI was obtained. This allows a
1398	   server to generate lists of back-links to documents, for interest,
1399	   logging, optimized caching, etc. It also allows obsolete or
1400	   mistyped links to be traced for maintenance. The format of the
1401	   field is:

1403	       Referer        = "Referer" ":" URI

1405	   Example:

1407	       Referer: http://info.cern.ch/hypertext/DataSources/Overview.html

1409	   If a partial URI is given, it should be interpreted relative to the
1410	   Request-URI.

1412	       Note: Because the source of a link may be considered private
1413	       information or may reveal an otherwise secure information
1414	       source, it is strongly recommended that the user be able to
1415	       select whether or not the Referer field is sent. For
1416	       example, a browser client could have a toggle switch for
1417	       browsing openly/anonymously, which would respectively
1418	       enable/disable the sending of Referer and From information.

1420	5.4.10 User-Agent

1422	   The User-Agent field contains information about the user agent
1423	   originating the request. This is for statistical purposes, the
1424	   tracing of protocol violations, and automated recognition of user
1425	   agents for the sake of tailoring responses to avoid particular user
1426	   agent limitations. Although it is not required, user agents should
1427	   always include this field with requests. The field can contain
1428	   multiple tokens specifying the product name, with an optional slash
1429	   and version designator, and other products which form a significant
1430	   part of the user agent. By convention, the products are listed in
1431	   order of their significance for identifying the application.

1433	       User-Agent      = "User-Agent" ":" 1*( product )

1435	       product         = token ["/" product-version]
1436	       product-version = token

1438	   Example:

1440	       User-Agent: CERN-LineMode/2.15 libwww/2.17b3

1442	   Product tokens should be short and to the point -- use of this field
1443	   for advertizing or other non-essential information is explicitly
1444	   deprecated and will be considered as non-conformance to the
1445	   protocol. Although any token character may appear in a
1446	   product-version, this token should only be used for a version
1447	   identifier (i.e., successive versions of the same product should
1448	   only differ in the product-version portion of the product value).
1449	   The User-Agent field may include additional information within
1450	   comments that are not part of the value of the field.

1452	       Note: Some current proxy applications append their product
1453	       information to the list in the User-Agent field. This is no
1454	       longer recommended, since it makes machine interpretation of
1455	       these fields ambiguous. Instead, proxies should use the
1456	       Forwarded header described in Section 4.3.2.

1458	6.  Response

1460	   After receiving and interpreting a request message, a server
1461	   responds in the form of an HTTP response message.

1463	       Response       = Simple-Response | Full-Response

1465	       Simple-Response= [ Entity-Body ]

1467	       Full-Response  = Status-Line               ; Section 6.1
1468	                        *( General-Header         ; Section 4.3
1469	                        |  Response-Header        ; Section 6.3
1470	                        |  Entity-Header )        ; Section 7.1
1471	                        CRLF
1472	                        [ Entity-Body ]           ; Section 7.2

1474	   A Simple-Response should only be sent in response to an HTTP/0.9
1475	   Simple-Request or if the server only supports the more limited
1476	   HTTP/0.9 protocol. If a client sends an HTTP/1.0 Full-Request and
1477	   receives a response that does not begin with a Status-Line, it
1478	   should assume that the response is a Simple-Response and parse it
1479	   accordingly. Note that the Simple-Response consists only of the
1480	   entity body and is terminated by the server closing the connection.

1482	6.1  Status-Line

1484	   The first line of a Full-Response message is the Status-Line,
1485	   consisting of the protocol version followed by a numeric status
1486	   code and its associated textual phrase, with each element separated
1487	   by SP characters. No CR or LF is allowed except in the final CRLF
1488	   sequence.

1490	       Status-Line = HTTP-Version SP Status-Code SP Reason-Phrase CRLF

1492	   Since a status line always begins with the protocol version's

1494	       "HTTP/" 1*DIGIT "." 1*DIGIT

1496	   (e.g. "HTTP/1.0"), the presence of that expression is considered
1497	   sufficient to differentiate a Full-Response from a Simple-Response.
1498	   Although the Simple-Response format may allow such an expression to
1499	   occur at the beginning of an entity body (and thus cause a
1500	   misinterpretation of the message if it was given in response to a
1501	   Full-Request), the likelihood of such an occurrence is negligible.

1503	6.2  Status Codes and Reason Phrases

1505	   The Status-Code element is a 3-digit integer result code of the
1506	   attempt to understand and satisfy the request. The Reason-Phrase is
1507	   intended to give a short textual description of the Status-Code.
1508	   The Status-Code is intended for use by automata and the Reason-
1509	   Phrase is intended for the human user. The client is not required
1510	   to examine or display the Reason-Phrase.

1512	   The first digit of the Status-Code defines the class of response.
1513	   The last two digits do not have any categorization role. There are
1514	   5 values for the first digit:

1516	      o 1xx: Informational - Not used, but reserved for future use

1518	      o 2xx: Success       - The action was successfully received,
1519	                             understood, and accepted.

1521	      o 3xx: Redirection   - Further action must be taken in order to
1522	                             complete the request

1524	      o 4xx: Client Error  - The request contains bad syntax or cannot
1525	                             be fulfilled

1527	      o 5xx: Server Error  - The server failed to fulfill an apparently
1528	                             valid request

1530	   The individual values of the numeric status codes defined for
1531	   HTTP/1.0, and an example set of corresponding Reason-Phrase's, are
1532	   presented below. The reason phrases listed here are only
1533	   recommended -- they may be replaced by local equivalents without
1534	   affecting the protocol.

1536	       Status-Code    = "200"   ; OK
1537	                      | "201"   ; Created
1538	                      | "202"   ; Accepted
1539	                      | "203"   ; Provisional Information
1540	                      | "204"   ; No Content
1541	                      | "300"   ; Multiple Choices
1542	                      | "301"   ; Moved Permanently
1543	                      | "302"   ; Moved Temporarily
1544	                      | "303"   ; Method
1545	                      | "304"   ; Not Modified
1546	                      | "400"   ; Bad Request
1547	                      | "401"   ; Unauthorized
1548	                      | "402"   ; Payment Required
1549	                      | "403"   ; Forbidden
1550	                      | "404"   ; Not Found
1551	                      | "405"   ; Method Not Allowed
1552	                      | "406"   ; None Acceptable
1553	                      | "407"   ; Proxy Authentication Required
1554	                      | "408"   ; Request Timeout
1555	                      | "409"   ; Conflict
1556	                      | "410"   ; Gone
1557	                      | "500"   ; Internal Server Error
1558	                      | "501"   ; Not Implemented
1559	                      | "502"   ; Bad Gateway
1560	                      | "503"   ; Service Unavailable
1561	                      | "504"   ; Gateway Timeout
1562	                      | extension-code

1564	       extension-code = 3DIGIT

1566	       Reason-Phrase  = token *( SP token )

1568	   HTTP status codes are extensible and should be registered with the
1569	   IANA. HTTP applications are not required to understand the meaning
1570	   of all registered status codes, though such understanding is
1571	   obviously desirable. However, applications are required to
1572	   understand the class of any status code (as indicated by the first
1573	   digit) and to treat the response as being equivalent to the x00
1574	   status code of that class. For example, if an unknown status code
1575	   of 421 is received by the client, it can safely assume that there
1576	   was something wrong with its request and treat the response as if
1577	   it had received a 400 status code. In such cases, user agents are
1578	   encouraged to present the entity returned with the response to the
1579	   user, since that entity is likely to include human-readable
1580	   information which will explain the unusual status.

1582	   Each Status-Code is described below, including a description of
1583	   which method(s) it can follow and any metainformation required in
1584	   the response.

1586	6.2.1 Successful 2xx

1588	   This class of status codes indicates that the client's request was
1589	   successfully received, understood, and accepted.

1591	   200 OK

1593	      o Following:        any method
1594	      o Required headers: none

1596	   The request has been fulfilled and an entity corresponding to the
1597	   requested resource is being sent in the response. If the HEAD
1598	   method was used, the response should only contain the Entity-Header
1599	   information and no Entity-Body.

1601	   201 Created

1603	      o Following:        any method that may request a new resource
1604	                          be created
1605	      o Required headers: URI-header

1607	   The request has been fulfilled and resulted in a new resource being
1608	   created. The newly created resource can be referenced by the URI(s)
1609	   returned in the URI-header field of the response. The origin server
1610	   is encouraged, but not obliged, to actually create the resource
1611	   before using this Status-Code. If the action cannot be carried out
1612	   immediately, or within a clearly defined timeframe, the server
1613	   should respond with "202 Accepted" instead.

1615	   Of the methods defined by this specification, only PUT and POST can
1616	   create a resource.

1618	   202 Accepted

1620	      o Following:        any method
1621	      o Required headers: none

1623	   The request has been accepted for processing, but the processing
1624	   has not been completed. The request may or may not eventually be
1625	   acted upon, as it may be disallowed when processing actually takes
1626	   place. There is no facility for re-sending a status code from an
1627	   asynchronous operation such as this.

1629	   The "202 Accepted" response is intentionally non-committal. Its
1630	   purpose is to allow a server to accept a request for some other
1631	   process (perhaps a batch-oriented process that is only run once per
1632	   day) without requiring that the user agent's connection to the
1633	   server persist until the process is completed. The entity returned
1634	   with this response should include an indication of the request's
1635	   current status and either a pointer to a status monitor or some
1636	   estimate of when the user can expect the request to be enacted.

1638	   203 Provisional Information

1640	      o Following:        GET, HEAD
1641	      o Required headers: none

1643	   The returned metainformation in the Entity-Header is not the
1644	   definitive set as available from the origin server, but is gathered
1645	   from a local or a third-party copy. The set presented may be a
1646	   subset or superset of the original version. For example, including
1647	   local annotation information about the resource may result in a
1648	   superset of the metainformation known by the origin server.

1650	   204 No Content

1652	      o Following:        any method
1653	      o Required headers: none

1655	   The server has fulfilled the request but there is no new
1656	   information to send back. If the client is a user agent, it should
1657	   not change its document view. This response is primarily intended
1658	   to allow input for scripts or other actions to take place without
1659	   causing a change to the user agent's current document view.

1661	6.2.2 Redirection 3xx

1663	   This class of status code indicates that further action needs to be
1664	   taken by the client in order to fulfill the request. The action
1665	   required can sometimes be carried out by the client without
1666	   interaction with the user, but it is strongly recommended that this
1667	   only takes place if the method used in the request is idempotent
1668	   (GET or HEAD).

1670	   300 Multiple Choices

1672	      o Following:        any method
1673	      o Required headers: none

1675	   The requested resource is available at one or more locations and a
1676	   preferred location could not be determined via content negotiation.
1677	   Unless it was a HEAD request, the response must include an entity
1678	   containing a formatted list of resource characteristics and
1679	   locations from which the user agent can choose the one most
1680	   appropriate. The entity format is specified by the media type given
1681	   in the Content-Type header field. Depending upon the format and the
1682	   capabilities of the user agent, selection of the most appropriate
1683	   choice may be performed automatically.

1685	   301 Moved Permanently

1687	      o Following:        any method
1688	      o Required headers: URI-header, Location

1690	   The requested resource has been assigned a new permanent URI and
1691	   any future references to this resource must be done using the
1692	   returned URI. Clients with link editing capabilities are encouraged
1693	   to automatically relink references to the Request-URI to the new
1694	   reference returned by the server, where possible.

1696	   It is possible for the server to send this status code in response
1697	   to a request using the PUT, POST, or DELETE methods. However, as
1698	   this might change the conditions under which the request was
1699	   issued, the user agent must not automatically redirect the request
1700	   unless it can be confirmed by the user.

1702	   302 Moved Temporarily

1704	      o Following:        any method
1705	      o Required headers: URI-header, Location

1707	   The requested resource resides temporarily under a different URI.
1708	   Since the redirection may be altered on occasion, the client should
1709	   on future requests from the user continue to use the original
1710	   Request-URI and not the URI returned in the URI-header field and
1711	   Location fields.

1713	   It is possible for the server to send this status code in response
1714	   to a request using the PUT, POST, or DELETE methods. However, as
1715	   this might change the conditions under which the request was
1716	   issued, the user agent must not automatically redirect the request
1717	   unless it can be confirmed by the user.

1719	   303 Method

1721	   This code is obsolete.

1723	   304 Not Modified

1725	      o Following:        conditional GET
1726	      o Required headers: none

1728	   If the client has performed a conditional GET request and access is
1729	   allowed, but the document has not been modified since the date and
1730	   time specified in the If-Modified-Since field, the server shall
1731	   respond with this status code and not send an Entity-Body to the
1732	   client. Header fields contained in the response should only include
1733	   information which is relevant to cache managers and which may have
1734	   changed independently of the entity's Last-Modified date. Examples
1735	   of relevant header fields include: Date, Server, and Expires.

1737	6.2.3 Client Error 4xx

1739	   The 4xx class of status codes is intended for cases in which the
1740	   client seems to have erred. If the client has not completed the
1741	   request when a 4xx code is received, it should immediately cease
1742	   sending data to the server. Except when responding to a HEAD
1743	   request, the server is encouraged to include an entity containing
1744	   an explanation of the error situation, and whether it is a
1745	   temporary or permanent condition. These status codes are applicable
1746	   to any request method.

1748	   400 Bad Request

1750	   The request had bad syntax or was inherently impossible to be
1751	   satisfied. The client is discouraged from repeating the request
1752	   without modifications.

1754	   401 Unauthorized

1756	      o Required headers: WWW-Authenticate

1758	   The request requires user authentication. The response must include
1759	   a WWW-Authenticate header field (Section 6.3.4) containing a
1760	   challenge applicable to the requested resource. The client may
1761	   repeat the request with a suitable Authorization header field.
1762	   HTTP access authentication is explained in Section 10.

1764	   402 Payment Required

1766	   This code is not currently supported, but is reserved for future
1767	   use.

1769	   403 Forbidden

1771	   The request is forbidden because of some reason that remains
1772	   unknown to the client. Authorization will not help and the request
1773	   should not be repeated. This status code can be used if the server
1774	   does not want to make public why the request cannot be fulfilled.

1776	   404 Not Found

1778	   The server has not found anything matching the Request-URI. No
1779	   indication is given of whether the condition is temporary or
1780	   permanent. If the server does not wish to make this information
1781	   available to the client, the status code "403 Forbidden" can be
1782	   used instead. The "410 Gone" status code should be used if the
1783	   server knows (through some internally configurable method) that an
1784	   old resource is permanently unavailable and has no forwarding
1785	   address.

1787	   405 Method Not Allowed

1789	      o Required headers: Allow

1791	   The method specified in the Request-Line is not allowed for the
1792	   resource identified by the Request-URI. The response must include
1793	   an Allow header containing a list of valid method's for the
1794	   requested resource.

1796	   406 None Acceptable

1798	      o Required headers: Content-*, where applicable to the Request-URI

1800	   The server has found a resource matching the Request-URI, but not
1801	   one that satisfies the conditions identified by the Accept and
1802	   Accept-Encoding request headers. The response must include (when
1803	   applicable) the Content-Type, Content-Encoding, and Content-
1804	   Language of the resource, and is encouraged to include the
1805	   resource's complete metainformation. No Entity-Body can be included
1806	   in the response.

1808	   407 Proxy Authentication Required

1810	   This code is reserved for future use. It is similar to
1811	   "401 Unauthorized", but indicates that the client must first
1812	   authenticate itself with the proxy. HTTP/1.0 does not provide a
1813	   means for proxy authentication -- this feature will be available in
1814	   future versions.

1816	   408 Request Timeout

1818	   The client did not produce a request within the time that the
1819	   server was prepared to wait. The client may repeat the request
1820	   without modifications at any later time.

1822	   409 Conflict

1824	   The request could not be completed due to a conflict with the
1825	   current state of the resource. This code is only allowed in
1826	   situations where it is expected that the user may be able to
1827	   resolve the conflict and resubmit the request. The response body
1828	   should include enough information for the user to recognize the
1829	   source of the conflict. Ideally, the response entity would include
1830	   enough information for the user (or user-agent) to fix the problem;
1831	   however, that may not be possible and is not required.

1833	   Conflicts are most likely to occur in response to a PUT request. If
1834	   versioning is being used and the entity being PUT includes changes
1835	   to a resource which conflict with those made by an earlier (third-
1836	   party) request, the server may use the "409 Conflict" response to
1837	   indicate that it can't complete the PUT. In this case, the response
1838	   entity may contain a list of the differences between the two
1839	   versions.

1841	   410 Gone

1843	   The requested resource is no longer available at the server and no
1844	   forwarding address is known. This condition should be considered
1845	   permanent. Clients with link editing capabilities are encouraged to
1846	   delete references to the Request-URI (after user approval). If the
1847	   server does not know (or has no facility to determine) whether or
1848	   not the condition is permanent, the status code "404 Not Found" can
1849	   be used instead.

1851	   The "410 Gone" response is primarily intended to assist the task of
1852	   web maintenance by notifying the recipient that the resource is
1853	   intentionally unavailable and that the server owners desire that
1854	   remote links to that resource be removed. Such an event is common
1855	   for limited-time, promotional services and for resources belonging
1856	   to individuals no longer working at the server's site. It is not
1857	   necessary to mark all permanently unavailable resources as "gone"
1858	   or to keep the mark for any length of time -- that is left to the
1859	   discretion of the server owner.

1861	6.2.4 Server Errors 5xx

1863	   Response status codes beginning with the digit "5" indicate cases
1864	   in which the server is aware that it has erred or is incapable of
1865	   performing the request. If the client has not completed the request
1866	   when a 5xx code is received, it should immediately cease sending
1867	   data to the server. Except when responding to a HEAD request, the
1868	   server is encouraged to include an entity containing an explanation
1869	   of the error situation, and whether it is a temporary or permanent
1870	   condition. These response codes are applicable to any request
1871	   method and there are no required header fields.

1873	   500 Internal Server Error

1875	   The server encountered an unexpected condition which prevented it
1876	   from fulfilling the request.

1878	   501 Not Implemented

1880	   The server does not support the functionality required to fulfill
1881	   the request. This is the appropriate response when the server does
1882	   not recognize the request method and is not capable of supporting
1883	   it for any resource.

1885	   502 Bad Gateway

1887	   The server received an invalid response from the gateway or
1888	   upstream server it accessed in attempting to complete the request.

1890	   503 Service Unavailable

1892	   The server is currently unable to handle the request due to a
1893	   temporary overloading or maintenance of the server. The implication
1894	   is that this is a temporary condition which will be alleviated
1895	   after some delay. If known, the length of the delay may be
1896	   indicated in a Retry-After header. If no Retry-After is given, the
1897	   client should handle the response as it would a "500 Internal
1898	   Server Error".

1900	       Note: The presence of the 503 status code does not imply
1901	       that a server must use it when becoming overloaded. Some
1902	       servers may wish to simply refuse the connection.

1904	   504 Gateway Timeout

1906	   The server did not receive a timely response from the gateway or
1907	   upstream server it accessed in attempting to complete the request.

1909	6.3  Response Header Fields

1911	   The response header fields allow the server to pass additional
1912	   information about the response which cannot be placed in the Status-
1913	   Line. These header fields are not intended to give information
1914	   about an Entity-Body returned in the response, but about the server
1915	   itself.

1917	       Response-Header= Public
1918	                      | Retry-After
1919	                      | Server
1920	                      | WWW-Authenticate
1921	                      | extension-header

1923	   Although additional response header fields may be implemented via
1924	   the extension mechanism, applications which do not recognize those
1925	   fields should treat them as Entity-Header fields.

1927	6.3.1 Public

1929	   The Public header field lists the set of non-standard methods
1930	   supported by the server. The purpose of this field is strictly to
1931	   inform the recipient of the capabilities of the server regarding
1932	   unusual methods. The methods listed may or may not be applicable to
1933	   the Request-URI; the Allow header field (Section 7.1.1) should be
1934	   used to indicate methods allowed for a particular URI. This does
1935	   not prevent a client from trying other methods. The field value
1936	   should not include the methods predefined for HTTP/1.0 in Section
1937	   5.2.

1939	       Public         = "Public" ":" 1#method

1941	   Example of use:

1943	       Public: OPTIONS, MGET, MHEAD

1945	   This header field applies only to the current connection. If the
1946	   response passes through a proxy, the proxy must either remove the
1947	   Public header field or replace it with one applicable to its own
1948	   capabilities.

1950	6.3.2 Retry-After

1952	   The Retry-After header field can be used with "503 Service
1953	   Unavailable" to indicate how long the service is expected to be
1954	   unavailable to the requesting client. The value of this field can
1955	   be either an full HTTP-date or an integer number of seconds (in
1956	   decimal) after the time of the response.

1958	       Retry-After    = "Retry-After" ":" ( HTTP-date | delta-seconds )

1960	   Two examples of its use are

1962	       Retry-After: Wed, 14 Dec 1994 18:22:54 GMT

1964	       Retry-After: 120

1966	   In the latter example, the delay is 2 minutes.

1968	6.3.3 Server

1970	   The Server header field contains information about the software
1971	   being used by the origin server program handling the request. The
1972	   field is analogous to the User-Agent field and has the following
1973	   format:

1975	       Server         = "Server" ":" 1*( product )

1977	   Example:

1979	       Server: CERN/3.0 libwww/2.17

1981	   If the response is being forwarded through a proxy, the proxy
1982	   application must not add its data to the product list. Instead, it
1983	   should include a Forwarded field, as described in Section 4.3.2.

1985	6.3.4 WWW-Authenticate

1987	   The WWW-Authenticate header field must be included as part of a
1988	   "401 Unauthorized" response. The field value consists of a
1989	   challenge that indicates the authentication scheme and parameters
1990	   applicable to the Request-URI.

1992	       WWW-Authenticate        = "WWW-Authenticate" ":" challenge

1994	   The HTTP access authentication process is described in Section 10.

1996	7.  Entity

1998	   Full-Request and Full-Response messages may transfer an entity
1999	   within some requests and responses. An entity consists of Entity-
2000	   Header fields and (usually) an Entity-Body. In this section, both
2001	   sender and recipient refer to either the client or the server,
2002	   depending on who sends and who receives the entity.

2004	7.1  Entity Header Fields

2006	   This section specifies the format and semantics of the Entity-
2007	   Header fields. Entity-Header fields define optional metainformation
2008	   about the Entity-Body or, if no body is present, about the resource
2009	   identified by the request.

2011	       Entity-Header  = Allow
2012	                      | Content-Encoding
2013	                      | Content-Language
2014	                      | Content-Length
2015	                      | Content-Transfer-Encoding
2016	                      | Content-Type
2017	                      | Derived-From
2018	                      | Expires
2019	                      | Last-Modified
2020	                      | Link
2021	                      | Location
2022	                      | Title
2023	                      | URI-header
2024	                      | Version
2025	                      | extension-header

2027	       extension-header = HTTP-header

2029	   Each entity header field is explained in the subsections below.
2030	   Other header fields are allowed but cannot be assumed to be
2031	   recognizable by the recipient. Unknown header fields should be
2032	   ignored by the recipient and forwarded by proxies.

2034	       Note: It has been proposed that any HTML metainformation
2035	       element (allowed within the  element of a document) be
2036	       a valid candidate for an HTTP entity header. This document
2037	       defines two header fields, Link and Title, which are both
2038	       examples of this. Base will be used in future versions of
2039	       HTTP.

2041	7.1.1 Allow

2043	   The Allow header field lists the set of methods supported by the
2044	   resource identified by the Request-URI. The purpose of this field
2045	   is strictly to inform the recipient of valid methods associated
2046	   with the resource; it must be present in a response with status
2047	   code "405 Method Not Allowed".

2049	       Allow          = "Allow" ":" 1#method

2051	    Example of use:

2053	       Allow: GET, HEAD, PUT

2055	   This does not prevent a client from trying other methods. However,
2056	   it is recommended that the indications given by this field be
2057	   followed. This field has no default value; if left undefined, the
2058	   set of allowed methods is defined by the origin server at the time
2059	   of each request.

2061	   If a response passes through a proxy which does not understand one
2062	   or more of the methods indicated in the Allow header, the proxy
2063	   must not modify the Allow header; the user agent may have other
2064	   means of communicating with the origin server.

2066	7.1.2 Content-Encoding

2068	   The Content-Encoding header field is used as a modifier to the
2069	   media-type. When present, its value indicates what additional
2070	   encoding mechanism has been applied to the resource, and thus what
2071	   decoding mechanism must be applied in order to obtain the media-
2072	   type referenced by the Content-Type header field. The Content-
2073	   Encoding is primarily used to allow a document to be compressed
2074	   without losing the identity of its underlying media type.

2076	       Content-Encoding = "Content-Encoding" ":" encoding-mechanism

2078	   Encoding mechanisms are defined in Section 8.4. An example of its
2079	   use is

2081	       Content-Encoding: gzip

2083	   The Content-Encoding is a characteristic of the resource identified
2084	   by the Request-URI. Typically, the resource is stored with this
2085	   encoding and is only decoded before rendering or analogous usage.

2087	7.1.3 Content-Language

2089	   The Content-Language field describes the natural language(s) of the
2090	   intended audience for the enclosed entity. Note that this may not
2091	   be equivalent to all the languages used within the entity.

2093	       Content-Language = "Content-Language" ":" 1#language-tag

2095	   Language tags are defined in Section 8.2. The primary purpose of
2096	   Content-Language is to allow a selective consumer to identify and
2097	   differentiate resources according to the consumer's own preferred
2098	   language. Thus, if the body content is intended only for a Danish
2099	   audience, the appropriate field is

2101	       Content-Language: dk

2103	   If no Content-Language is specified, the default is that the
2104	   content is intended for all language audiences. This may mean that
2105	   the sender does not consider it to be specific to any natural
2106	   language, or that the sender does not know for which language it is
2107	   intended.

2109	   Multiple languages may be listed for content that is intended for
2110	   multiple audiences. For example, a rendition of the "Treaty of
2111	   Waitangi," presented simultaneously in the original Maori and
2112	   English versions, would call for

2114	       Content-Language: mi, en

2116	   However, just because multiple languages are present within an
2117	   entity does not mean that it is intended for multiple linguistic
2118	   audiences. An example would be a beginner's language primer, such
2119	   as "A First Lesson in Latin," which is clearly intended to be used
2120	   by an English audience. In this case, the Content-Language should
2121	   only include "en".

2123	   Content-Language may be applied to any media type -- it should not
2124	   be considered limited to textual documents.

2126	7.1.4 Content-Length

2128	   The Content-Length header field indicates the size of the Entity-
2129	   Body (in decimal number of octets) sent to the recipient or, in the
2130	   case of the HEAD method, the size of the Entity-Body that would
2131	   have been sent had the request been a GET.

2133	       Content-Length = "Content-Length" ":" 1*DIGIT

2135	   An example is

2137	       Content-Length: 3495

2139	   Although it is not required, applications are strongly encouraged
2140	   to use this field to indicate the size of the Entity-Body to be
2141	   transferred, regardless of the media type of the entity.

2143	   Any Content-Length greater than or equal to zero is a valid value.
2144	   Section 7.2.2 describes how to determine the length of an Entity-
2145	   Body if a Content-Length is not given.

2147	       Note: The meaning of this field is significantly different
2148	       from the corresponding definition in MIME, where it is an
2149	       optional field used within the "message/external-body"
2150	       content-type. In HTTP, it should be used whenever the
2151	       entity's length can be determined prior to being transferred.

2153	7.1.5 Content-Transfer-Encoding

2155	   The Content-Transfer-Encoding (CTE) header indicates what (if any)
2156	   type of transformation has been applied to the entity in order to
2157	   safely transfer it between the sender and the recipient. This
2158	   differs from the Content-Encoding in that the CTE is a property of
2159	   the message, not of the original resource.

2161	       Content-Transfer-Encoding
2162	            = "Content-Transfer-Encoding" ":" transfer-encoding

2164	   Transfer encodings are defined in Section 8.5. Because all HTTP
2165	   transactions take place on an 8-bit clean connection, the default
2166	   Content-Transfer-Encoding for all messages is binary. However, HTTP
2167	   may be used to transfer MIME messages which already have a defined
2168	   CTE. An example is:

2170	       Content-Transfer-Encoding: quoted-printable

2172	   Many older HTTP/1.0 applications do not understand the Content-
2173	   Transfer-Encoding header. However, since it may appear in any MIME
2174	   message (i.e. entities retrieved via a gateway to a MIME-conformant
2175	   protocol), future HTTP/1.0 applications are required to understand
2176	   it upon receipt. Gateways are the only HTTP applications that would
2177	   generate a CTE.

2179	7.1.6 Content-Type

2181	   The Content-Type header field indicates the media type of the
2182	   Entity-Body sent to the recipient or, in the case of the HEAD
2183	   method, the media type that would have been sent had the request
2184	   been a GET.

2186	       Content-Type = "Content-Type" ":" media-type

2188	   Media types are defined in Section 8.1. An example of the field is

2190	       Content-Type: text/html; charset=ISO-8859-4

2192	   The Content-Type header field has no default value. Further
2193	   discussion of methods for identifying the media type of an entity
2194	   is provided in Section 7.2.1.

2196	7.1.7 Derived-From

2198	   The Derived-From header field can be used to indicate the version
2199	   tag of the resource from which the enclosed entity was derived
2200	   before modifications were made by the sender. This field is used to
2201	   help manage the process of merging successive changes to a
2202	   resource, particularly when such changes are being made in parallel
2203	   and from multiple sources.

2205	       Derived-From   = "Derived-From" ":" version-tag

2207	   An example use of the field is:

2209	       Derived-From: 2.1.1

2211	   A longer example of version control is included in Appendix C. The
2212	   Derived-From field is required for PUT requests if the entity being
2213	   put was previously retrieved from the same URI and a Version header
2214	   was included with the entity when it was last retrieved.

2216	7.1.8 Expires

2218	   The Expires field gives the date/time after which the entity should
2219	   be considered stale. This allows information providers to suggest
2220	   the volatility of the resource. Caching clients (including proxies)
2221	   must not cache this copy of the resource beyond the date given,
2222	   unless its status has been updated by a later check of the origin
2223	   server. The presence of an Expires field does not imply that the
2224	   original resource will change or cease to exist at, before, or
2225	   after that time. However, information providers that know (or even
2226	   suspect) that a resource will change by a certain date are strongly
2227	   encouraged to include an Expires header with that date. The format
2228	   is an absolute date and time as defined by HTTP-date in Section 3.3.

2230	       Expires        = "Expires" ":" HTTP-date

2232	   An example of its use is

2234	       Expires: Thu, 01 Dec 1994 16:00:00 GMT

2236	   The Expires field has no default value. If the date given is equal
2237	   to or earlier than the value of the Date header, the recipient must
2238	   not cache the enclosed entity. If a resource is dynamic by nature,
2239	   as is the case with many data-producing processes, copies of that
2240	   resource should be given an appropriate Expires value which
2241	   reflects that dynamism.

2243	   The Expires field cannot be used to force a user agent to refresh
2244	   its display or reload a resource; its semantics apply only to
2245	   caching mechanisms, and such mechanisms need only check a
2246	   resource's expiration status when a new request for that resource
2247	   is initiated.

2249	       Note: Applications are encouraged to be tolerant of bad or
2250	       misinformed implementations of the Expires header. In
2251	       particular, recipients may wish to recognize a delta-seconds
2252	       value (any decimal integer) as representing the number of
2253	       seconds after receipt of the message that its contents
2254	       should be considered expired. Likewise, a value of zero (0)
2255	       or an invalid date format may be considered equivalent to an
2256	       "expires immediately." Although these values are not
2257	       legitimate for HTTP/1.0, a robust implementation is always
2258	       desirable.

2260	7.1.9 Last-Modified

2262	   The Last-Modified header field indicates the date and time at which
2263	   the sender believes the resource was last modified. The exact
2264	   semantics of this field are defined in terms of how the receiver
2265	   should interpret it:  if the receiver has a copy of this resource
2266	   which is older than the date given by the Last-Modified field, that
2267	   copy should be considered stale.

2269	       Last-Modified  = "Last-Modified" ":" HTTP-date

2271	   An example of its use is

2273	       Last-Modified: Tue, 15 Nov 1994 12:45:26 GMT

2275	   The exact meaning of this header field depends on the
2276	   implementation of the sender and the nature of the original
2277	   resource. For files, it may be just the file system last-mod date.
2278	   For entities with dynamically included parts, it may be the most
2279	   recent of the set of last-mod dates of its component parts. For
2280	   database gateways, it may be the last-update timestamp of the
2281	   record. For virtual objects, it may be the last time the internal
2282	   state changed. In any case, the recipient should only know (and
2283	   care) about the result -- whatever gets stuck in the Last-Modified
2284	   field -- and not worry about how that result was obtained.

2286	7.1.10 Link

2288	   The Link header provides a means for describing a relationship
2289	   between the entity and some other resource. An entity may include
2290	   multiple Link values. Links at the metainformation level typically
2291	   indicate relationships like hierarchical structure and navigation
2292	   paths. The Link field is semantically equivalent to the 
2293	   element in HTML [4].

2295	       Link           = "Link" ":" 1#("<" URI ">"
2296	                        [ ";" "rel" "=" relation ]
2297	                        [ ";" "rev" "=" relation ]
2298	                        [ ";" "title" "=" quoted-string ] )

2300	       relation       = sgml-name

2302	       sgml-name      = ALPHA *( ALPHA | DIGIT | "." | "-" )

2304	   Relation values are not case-sensitive and may be extended within
2305	   the constraints of the sgmlname syntax. There are no predefined
2306	   link relationship values for HTTP/1.0. The title parameter may be
2307	   used to label the destination of a link such that it can be used as
2308	   identification within a human-readable menu. Examples of usage
2309	   include:

2311	       Link: ; rel="Previous"

2313	       Link: ; rev="Made"; title="Tim Berners-Lee"

2315	   The first example indicates that the entity is previous to chapter2
2316	   in a logical navigation path. The second indicates that the
2317	   publisher of the resource is identified by the given e-mail address.

2319	7.1.11 Location

2321	   The Location header field is an earlier form of the URI-header and
2322	   is considered obsolete. However, HTTP/1.0 applications should
2323	   continue to support the Location header in order to properly
2324	   interface with older applications. The purpose of Location is
2325	   identical to that of the URI-header, except that no variants can be
2326	   specified and only one absolute location URL is allowed.

2328	       Location       = "Location" ":" URI

2330	   An example is

2332	       Location: http://info.cern.ch/hypertext/WWW/NewLocation.html

2334	7.1.12 Title

2336	   The Title header field indicates the title of the entity

2338	       Title          = "Title" ":" *text

2340	   An example of the field is

2342	       Title: Hypertext Transfer Protocol -- HTTP/1.0

2344	   This field is to be considered isomorphic with the  element
2345	   in HTML [4].

2347	7.1.13 URI

2349	   The URI-header field may contain one or more Universal Resource
2350	   Identifiers (URIs) by which the resource origin of the entity can
2351	   be identified. There is no guarantee that the resource can be
2352	   accessed using the URI(s) specified. This field is required for the
2353	   201, 301, and 302 response messages, and may be included in any
2354	   message that contains resource metainformation.

2356	       URI-header     = "URI" ":" 1#( "<" URI ">" [ ";" vary ] )

2358	       vary           = "vary" "=" <"> 1#vary-dimension <">
2359	       vary-dimension = "type" | "language" | "version" | "encoding"
2360	                      | "charset" | "user-agent" | extension-vary

2362	       extension-vary = token

2364	   Any URI specified in this field can be either absolute or relative
2365	   to the URI given in the Request-Line. The URI-header improves upon
2366	   the Location header field described in Section 7.1.11. For
2367	   backwards compatibility with older clients, servers are encouraged
2368	   to include both header fields in 301 and 302 responses.

2370	   The URI-header may also be used by a client performing a POST
2371	   request to suggest a URI for the new entity. Whether or not the
2372	   suggested URI is used is entirely up to the server to decide. In
2373	   any case, the server's response must include the actual URI(s) of
2374	   the new resource if one is successfully created (status 201).

2376	   If a URI refers to a set of variants, then the dimensions of that
2377	   variance must be given with a vary parameter. One example is:

2379	       URI: <http://info.cern.ch/hypertext/WWW/TheProject.multi>;
2380	            vary="type,language"

2382	   which indicates that the URI covers a group of entities that vary
2383	   in media type and natural language. A request for that URI will
2384	   result in a response that depends upon the client's request headers
2385	   for Accept and Accept-Language. Similar dimensions exist for the
2386	   Accept-Encoding, Accept-Charset, Version, and User-Agent header
2387	   fields, as demonstrated in the following example.

2389	       URI: <TheProject.ps>;vary="encoding,version",
2390	            <TheProject.html>;vary="user-agent,charset,version"

2392	   The vary parameter has an important effect on cache management,
2393	   particularly for caching proxies which service a diverse set of
2394	   user agents. Since the response to one user agent may differ from
2395	   the response to a second user agent if the two agents have
2396	   differing request profiles, a caching proxy must keep track of the
2397	   content metainformation for resources with varying dimensions.
2398	   Thus, the vary parameter tells the caching proxy what entity
2399	   headers must be part of the key for caching that URI. When the
2400	   caching proxy gets a request for that URI, it must forward the
2401	   request toward the origin server if the request profile includes a
2402	   variant dimension that has not already been cached.

2404	7.1.14 Version

2406	   The Version field defines the version tag associated with a
2407	   rendition of an evolving entity. Together with the Derived-From
2408	   field described in Section 7.1.7, it allows a group of people to
2409	   work simultaneously on the creation of a work as an iterative
2410	   process. The field should be used to allow evolution of a
2411	   particular work along a single path. It should not be used to
2412	   indicate derived works or renditions in different representations.

2414	       Version        = "Version" ":" version-tag

2416	       version-tag    = token | quoted-string

2418	   Examples of the Version field include:

2420	       Version: 2.1.2
2421	       Version: "Fred 19950116-12:26:48"
2422	       Version: 2.5a4-omega7

2424	   The version tag should be considered opaque to all parties but the
2425	   origin server. A user agent can request a particular version of an
2426	   entity by including its tag in a Version header as part of the
2427	   request. Similarly, a user agent may suggest a value for the
2428	   version of an entity transferred via a PUT or POST request.
2429	   However, only the origin server can reliably assign or increment
2430	   the version tag of an entity.

2432	7.2  Entity Body

2434	   The entity-body (if any) sent with an HTTP/1.0 request or response
2435	   is in a format and encoding defined by the Entity-Header fields.

2437	       Entity-Body    = *OCTET

2439	   An entity-body is included with a request message only when the
2440	   request method calls for one. This specification defines two
2441	   request methods, "POST" and "PUT", that allow an entity-body. In
2442	   general, the presence of an entity-body in a request is signaled by
2443	   the inclusion of a Content-Length and/or Content-Transfer-Encoding
2444	   header field in the request message headers.

2446	       Note: Most current implementations of the POST and PUT
2447	       methods require a valid Content-Length header field. This
2448	       can cause problems for some systems that do not know the
2449	       size of the entity-body before transmission. Experimental
2450	       implementations (and future versions of HTTP) may use a
2451	       packetized Content-Transfer-Encoding to obviate the need for
2452	       a Content-Length.

2454	   For response messages, whether or not an entity-body is included
2455	   with a message is dependent on both the request method and the
2456	   response code. All responses to the HEAD request method must not
2457	   include a body, even though the presence of Content header fields
2458	   may lead one to believe they should. Similarly, the responses "204
2459	   No Content", "304 Not Modified", and "406 None Acceptable" must not
2460	   include a body.

2462	7.2.1 Type

2464	   When an Entity-Body is included with a message, the data type of
2465	   that body is determined via the header fields Content-Type,
2466	   Content-Encoding, and Content-Transfer-Encoding. These define a
2467	   three-layer, ordered encoding model:

2469	       entity-body <-
2470	          Content-Transfer-Encoding( Content-Encoding( Content-Type ) )

2472	   The default for both encodings is none (i.e., the identity
2473	   function). A Content-Type specifies the media type of the
2474	   underlying data. A Content-Encoding may be used to indicate an
2475	   additional encoding mechanism applied to the type (usually for the
2476	   purpose of data compression) that is a property of the resource
2477	   requested. A Content-Transfer-Encoding may be applied by a
2478	   transport agent to ensure safe and/or proper transfer of the
2479	   message. Note that the Content-Transfer-Encoding is a property of
2480	   the message, not of the resource.

2482	   The Content-Type header field has no default value. If and only if
2483	   the media type is not given by a Content-Type header (as is always
2484	   the case for Simple-Response messages), the receiver may attempt to
2485	   guess the media type via inspection of its content and/or the name
2486	   extension(s) of the URL used to access the resource. If the media
2487	   type remains unknown, the receiver should treat it as type
2488	   "application/octet-stream".

2490	7.2.2 Length

2492	   When an Entity-Body is included with a message, the length of that
2493	   body may be determined in one of several ways. If a Content-Length
2494	   header field is present, its value in bytes (number of octets)
2495	   represents the length of the Entity-Body. Otherwise, the body
2496	   length is determined by the Content-Type (for types with an
2497	   explicit end-of-body delimiter), the Content-Transfer-Encoding (for
2498	   packetized encodings), or the closing of the connection by the
2499	   server. Note that the latter cannot be used to indicate the end of
2500	   a request body, since it leaves no possibility for the server to
2501	   send back a response.

2503	       Note: Some older servers supply an invalid Content-Length
2504	       when sending a document that contains server-side includes
2505	       dynamically inserted into the data stream. It must be
2506	       emphasized that this will not be tolerated by future
2507	       versions of HTTP. Unless the client knows that it is
2508	       receiving a response from a compliant server, it should not
2509	       depend on the Content-Length value being correct.

2511	8.  Content Parameters

2513	8.1  Media Types

2515	   HTTP uses Internet Media Types [15], formerly referred to as MIME
2516	   Content-Types [6], in order to provide open and extensible data
2517	   typing and type negotiation. For mail applications, where there is
2518	   no type negotiation between sender and receiver, it is reasonable
2519	   to put strict limits on the set of allowed media types. With HTTP,
2520	   however, user agents can identify acceptable media types as part of
2521	   the connection, and thus are allowed more freedom in the use of non-
2522	   registered types. The following grammar for media types is a
2523	   superset of that for MIME because it does not restrict itself to
2524	   the official IANA and x-token types.

2526	       media-type     = type "/" subtype *( ";" parameter )
2527	       type           = token
2528	       subtype        = token

2530	    Parameters may follow the type/subtype in the form of
2531	   attribute/value pairs.

2533	       parameter      = attribute "=" value
2534	       attribute      = token
2535	       value          = token | quoted-string

2537	   The type, subtype, and parameter attribute names are not case-
2538	   sensitive. Parameter values may or may not be case-sensitive,
2539	   depending on the semantics of the parameter name. No LWS is allowed
2540	   between the type and subtype, nor between an attribute and its
2541	   value.

2543	   If a given media-type value has been registered by the IANA, any
2544	   use of that value must be indicative of the registered data format.
2545	   Although HTTP allows the use of non-registered media types, such
2546	   usage must not conflict with the IANA registry. Data providers are
2547	   strongly encouraged to register their media types with IANA via the
2548	   procedures outlined in RFC 1590 [15].

2550	   All media-type's registered by IANA must be preferred over
2551	   extension tokens. However, HTTP does not limit conforming
2552	   applications to the use of officially registered media types, nor
2553	   does it encourage the use of an "x-" prefix for unofficial types
2554	   outside of explicitly short experimental use between consenting
2555	   applications.

2557	8.1.1 Canonicalization and Text Defaults

2559	   Media types are registered in a canonical form. In general, entity
2560	   bodies transferred via HTTP must be represented in the appropriate
2561	   canonical form prior to transmission. If the body has been encoded
2562	   via a Content-Encoding and/or Content-Transfer-Encoding, the data
2563	   must be in canonical form prior to that encoding. However, HTTP
2564	   modifies the canonical form requirements for media of primary type
2565	   "text" and for "application" types consisting of text-like records.

2567	   HTTP redefines the canonical form of text media to allow multiple
2568	   octet sequences to indicate a text line break. In addition to the
2569	   preferred form of CRLF, HTTP applications must accept a bare CR or
2570	   LF alone as representing a single line break in text media.
2571	   Furthermore, if the text media is represented in a character set
2572	   which does not use octets 13 and 10 for CR and LF respectively (as
2573	   is the case for some multi-byte character sets), HTTP allows the
2574	   use of whatever octet sequence(s) is defined by that character set
2575	   to represent the equivalent of CRLF, bare CR, and bare LF. It is
2576	   assumed that any recipient capable of using such a character set
2577	   will know the appropriate octet sequence for representing line
2578	   breaks within that character set.

2580	       Note: This interpretation of line breaks applies only to the
2581	       contents of an Entity-Body and only after any Content-
2582	       Transfer-Encoding and/or Content-Encoding has been removed.
2583	       All other HTTP constructs use CRLF exclusively to indicate a
2584	       line break. Encoding mechanisms define their own line break
2585	       requirements.

2587	   A recipient of an HTTP text entity should translate the received
2588	   entity line breaks to the local line break conventions before
2589	   saving the entity external to the application and its cache;
2590	   whether this translation takes place immediately upon receipt of
2591	   the entity, or only when prompted by the user, is entirely up to
2592	   the individual application.

2594	   HTTP also redefines the default character set for text media in an
2595	   entity body. If a textual media type defines a charset parameter
2596	   with a registered default value of "US-ASCII", HTTP changes the
2597	   default to be "ISO-8859-1". Since the character set ISO-8859-1 [19]
2598	   is a superset of USASCII [18], this has no effect upon the
2599	   interpretation of entity bodies which only contain octets within
2600	   the US-ASCII set (0 - 127). The presence of a charset parameter
2601	   value in a Content-Type header field overrides the default.

2603	   HTTP does not require that the character set of an entity body be
2604	   labelled as the lowest common denominator of the character codes
2605	   used within a document.

2607	8.1.2 Multipart Types

2609	   MIME provides for a number of "multipart" types -- encapsulations of
2610	   several entities within a single message's Entity-Body. The
2611	   multipart types registered by IANA [17] do not have any special
2612	   meaning for HTTP/1.0, though user agents may need to understand
2613	   each type in order to correctly interpret the purpose of each body-
2614	   part. Ideally, an HTTP user agent should follow the same or similar
2615	   behavior as a MIME user agent does upon receipt of a multipart type.

2617	   As in MIME [6], all multipart types share a common syntax and must
2618	   include a boundary parameter as part of the media type value. The
2619	   message body is itself a protocol element and must therefore use
2620	   only CRLF to represent line breaks between body-parts. Unlike in
2621	   MIME, multipart body-parts may contain HTTP header fields which are
2622	   significant to the meaning of that part.

2624	   A URI-header field (Section 7.1.13) should be included in the body-
2625	   part for each enclosed entity that can be identified by a URI.

2627	8.2  Language Tags

2629	   A language tag identifies a natural language spoken, written, or
2630	   otherwise conveyed by human beings for communication of information
2631	   to other human beings. Computer languages are explicitly excluded.
2632	   The HTTP/1.0 protocol uses language tags within the Accept-Language
2633	   and Content-Language header fields.

2635	   The syntax and registry of HTTP language tags is the same as that
2636	   defined by RFC 1766 [1]. In summary, a language tag is composed of
2637	   1 or more parts: A primary language tag and a (possibly empty)
2638	   series of subtags:

2640	        language-tag  = primary-tag *( "-" subtag )

2642	        primary-tag   = 1*8ALPHA
2643	        subtag        = 1*8ALPHA

2645	   Whitespace is not allowed within the tag and all tags are to be
2646	   treated as case insensitive. The namespace of language tags is
2647	   administered by the IANA. Example tags include:

2649	       en, en-US, en-cockney, i-cherokee, x-pig-latin

2651	   where any two-letter primary-tag is an ISO 639 language
2652	   abbreviation and any two-letter initial subtag is an ISO 3166
2653	   country code.

2655	       Note: Earlier versions of this document specified an
2656	       incomplete language tag, where values were limited to ISO
2657	       639 language abbreviations with an optional ISO 3166 country
2658	       code appended after an underscore ("_") or slash ("/")
2659	       character. This format was abandoned in favor of the
2660	       recently proposed standard for Internet protocols.

2662	8.3  Character Sets

2664	   HTTP uses the same definition of the term "character set" as that
2665	   described for MIME:

2667	        The term "character set" is used in this document to
2668	        refer to a method used with one or more tables to convert
2669	        a sequence of octets into a sequence of characters. Note
2670	        that unconditional conversion in the other direction is
2671	        not required, in that not all characters may be available
2672	        in a given character set and a character set may provide
2673	        more than one sequence of octets to represent a
2674	        particular character. This definition is intended to
2675	        allow various kinds of character encodings, from simple
2676	        single-table mappings such as US-ASCII to complex table
2677	        switching methods such as those that use ISO 2022's
2678	        techniques. However, the definition associated with a
2679	        MIME character set name must fully specify the mapping to
2680	        be performed from octets to characters. In particular,
2681	        use of external profiling information to determine the
2682	        exact mapping is not permitted.

2684	   Character sets are identified by case-insensitive tokens. The
2685	   complete set of allowed charset values are defined by the IANA
2686	   Character Set registry [17]. However, because that registry does
2687	   not define a single, consistent token for each character set, we
2688	   define here the preferred names for those character sets most
2689	   likely to be used with HTTP entities. This set of charset values
2690	   includes those registered by RFC 1521 [6] -- the US-ASCII [18] and
2691	   ISO8859 [19] character sets -- and other character set names
2692	   specifically recommended for use within MIME charset parameters.

2694	     charset = "US-ASCII"
2695	             | "ISO-8859-1" | "ISO-8859-2" | "ISO-8859-3"
2696	             | "ISO-8859-4" | "ISO-8859-5" | "ISO-8859-6"
2697	             | "ISO-8859-7" | "ISO-8859-8" | "ISO-8859-9"
2698	             | "ISO-2022-JP" | "ISO-2022-JP-2" | "ISO-2022-KR"
2699	             | "UNICODE-1-1" | "UNICODE-1-1-UTF-7" | "UNICODE-1-1-UTF-8"
2700	             | token

2702	   Although HTTP allows an arbitrary token to be used as a character
2703	   set value, any token that has a predefined value within the IANA
2704	   Character Set registry [17] must represent the character set
2705	   defined by that registry. Applications are encouraged, but not
2706	   required, to limit their use of character sets to those defined by
2707	   the IANA registry.

2709	8.4  Encoding Mechanisms

2711	   Encoding mechanism values are used to indicate an encoding
2712	   transformation that has been or can be applied to a resource.
2713	   Encoding mechanisms are primarily used to allow a document to be
2714	   compressed or encrypted without losing the identity of its
2715	   underlying media type. Typically, the resource is stored with this
2716	   encoding and is only decoded before rendering or analogous usage.

2718	       encoding-mechanism      = "gzip" | "compress" | token

2720	       Note: For historical reasons, HTTP/1.0 applications should
2721	       consider "x-gzip" and "x-compress" to be equivalent to
2722	       "gzip" and "compress", respectively.

2724	   All encoding-mechanism values are case-insensitive. HTTP/1.0 uses
2725	   encoding-mechanism values in the Accept-Encoding (Section 5.4.3)
2726	   and Content-Encoding (Section 7.1.2) header fields. Although the
2727	   value describes the encoding-mechanism, what is more important is
2728	   that it indicates what decoding mechanism will be required to
2729	   remove the encoding. Note that a single program may be capable of
2730	   decoding multiple encoding-mechanism formats. Two values are
2731	   defined by this specification:

2733	   gzip      An encoding format produced by the file compression program
2734	             "gzip" (GNU zip) developed by Jean-loup Gailly. This format
2735	             is typically a Lempel-Ziv coding (LZ77) with a 32 bit CRC.
2736	             Gzip is available from the GNU project at
2737	             <URL:ftp://prep.ai.mit.edu/pub/gnu/>.

2739	   compress  The encoding format produced by the file compression
2740	             program "compress". This format is an adaptive Lempel-Ziv-
2741	             Welch coding (LZW).

2743	8.5  Transfer Encodings

2745	   Transfer encoding values are used to indicate an encoding
2746	   transformation that has been, can be, or may need to be applied to
2747	   an Entity-Body in order to ensure safe transport through the
2748	   network. Transfer encodings are only used with entities destined
2749	   for or retrieved from MIME-conformant systems, and thus will rarely
2750	   occur in an HTTP/1.0 message. This differs from an encoding-
2751	   mechanism in that the transfer encoding is a property of the
2752	   message, not of the original resource.

2754	       transfer-encoding       = "binary" | "8bit" | "7bit"
2755	                               | "quoted-printable" | "base64"
2756	                               | token

2758	   All transfer-encoding values are case-insensitive. HTTP/1.0 uses
2759	   transfer-encoding values in the Accept-Encoding (Section 5.4.3) and
2760	   Content-Transfer-Encoding (Section 7.1.5) header fields.

2762	   The values "7bit", "8bit", and "binary" are used to indicate that
2763	   no transfer encoding has been performed. Instead, they describe the
2764	   sort of encoding that might be needed for transmission through an
2765	   unsafe transport system. Binary indicates that the body may contain
2766	   any set of octets. 8bit adds the restrictions that CR and LF
2767	   characters only occur as part of CRLF line separators, all lines
2768	   are short (less than 1000 octets), and no NULs (octet 0) are
2769	   present. 7bit adds a further restriction that all octets are 7-bit
2770	   US-ASCII characters.

2772	   The "quoted-printable" and "base64" values indicate that the
2773	   associated encoding (as defined in MIME [6]) has been applied to
2774	   the body. These encodings consist entirely of 7-bit US-ASCII
2775	   characters.

2777	9.  Content Negotiation

2779	   Content negotiation is an optional feature of the HTTP protocol. It
2780	   is designed to allow for preemptive selection of a preferred
2781	   content representation, within a single request-response round-trip,
2782	   and without intervention from the user. However, this may not
2783	   always be desirable for the user and is sometimes unnecessary for
2784	   the content provider. Implementors are encouraged to provide
2785	   mechanisms whereby the amount of preemptive content negotiation,
2786	   and the parameters of that negotiation, are configurable by the
2787	   user and server maintainer.

2789	   The first step in the negotiation algorithm is for the server to
2790	   determine whether or not there are any content variants for the
2791	   requested resource. Content variants may be in the form of multiple
2792	   preexisting entities or a set of dynamic conversion filters. If
2793	   there are no variant forms of the resource, the "negotiation" is
2794	   limited to whether or not that single media type is acceptable
2795	   under the constraints given by the Accept request header field
2796	   (if any).

2798	   If variants are available, those variants that are completely
2799	   unacceptable should be removed from consideration first.
2800	   Unacceptable variants include those with a Content-Encoding not
2801	   listed in an Accept-Encoding field, those with a character subset
2802	   (other than the default ISO-8859-1) not listed in an Accept-Charset
2803	   field, and those with a media type not within any of the media
2804	   ranges of an explicitly constrained Accept field (or listed with a
2805	   zero quality parameter).

2807	   If no acceptable variants remain at this point, the server should
2808	   respond with a "406 None Acceptable" response message. If more than
2809	   one variant remains, and at least one has a Content-Language within
2810	   those listed by an Accept-Language field, any variants which do not
2811	   match the language constraint are removed from further
2812	   consideration.

2814	   If multiple choices still remain, the selection is further winnowed
2815	   by calculating and comparing the relative quality of the available
2816	   media types. The calculated weights are normalized to a real number
2817	   in the range 0 through 1, where 0 is the minimum and 1 the maximum
2818	   value. The following parameters are included in the calculation:

2820	      q    The quality factor chosen by the user agent (and
2821	           configurable by the user) that represents the desirability
2822	           of that media type. In this case, desirability is usually a
2823	           measure of the clients ability to faithfully represent the
2824	           contents of that media type to the user. The value is in
2825	           the range [0,1], where the default value is 1.

2827	      qs   The quality factor, as chosen by the server (via some
2828	           unspecified mechanism), to represent the relative quality
2829	           of a particular variant representation of the source. For
2830	           example, a picture originally in JPEG form would have a
2831	           lower qs when translated to the XBM format, and much lower
2832	           qs when translated to an ASCII-art representation. Note,
2833	           however, that this is a function of the source -- an
2834	           original piece of ASCII-art may degrade in quality if it is
2835	           captured in JPEG form. The qs value has the same range and
2836	           default as q.

2838	      mxb  The maximum number of bytes in the Entity-Body accepted by
2839	           the client. The default value is mxb=undefined
2840	           (i.e. infinity).

2842	      bs   The actual number of bytes in the Entity-Body for a
2843	           particular source representation. This should equal the
2844	           value sent as Content-Length. The default value is 0.

2846	   The discrete mapping function is defined as:

2848	                        { if mxb=undefined, then (qs * q) }
2849	       Q(q,qs,mxb,bs) = { if mxb >= bs,     then (qs * q) }
2850	                        { if mxb <  bs,     then 0        }

2852	   The variants with a maximal value for the Q function represent the
2853	   preferred representation(s) of the entity. If multiple
2854	   representations exist for a single media type (as would be the case
2855	   if they only varied by Content-Encoding), then the smallest
2856	   representation (lowest bs) is preferred.

2858	   Finally, there may still be multiple choices available to the user.
2859	   If so, the server may either choose one (possibly at random) from
2860	   those available and respond with "200 OK", or it may respond with
2861	   "300 Multiple Choices" and include an entity describing the
2862	   choices. In the latter case, the entity should either be of type
2863	   "text/html' (such that the user can choose from among the choices
2864	   by following an exact link) or of some type that would allow the
2865	   user agent to perform the selection automatically (no such type is
2866	   available at the time of this writing).

2868	   The "300 Multiple Choices" response can be given even if the server
2869	   does not perform any winnowing of the representation choices via
2870	   the content negotiation algorithm described above. Furthermore, it
2871	   may include choices that were not considered as part of the
2872	   negotiation algorithm and resources that may be located at other
2873	   servers.

2875	10.  Access Authentication

2877	   HTTP provides a simple challenge-response authorization mechanism
2878	   which may be used by a server to challenge a client request and by
2879	   a client to provide authentication information. The mechanism uses
2880	   an extensible token to identify the authentication scheme, followed
2881	   by a comma-separated list of attribute-value pairs which carry the
2882	   parameters necessary for achieving authentication via that scheme.

2884	       auth-scheme    = "Basic" | token

2886	       auth-param     = token "=" quoted-string

2888	   The "401 Unauthorized" response message is used by an origin server
2889	   to challenge the authorization of a user agent. This response must
2890	   include a WWW-Authenticate header field containing the challenge
2891	   applicable to the requested resource.

2893	       challenge      = auth-scheme 1*LWS realm [ "," 1#auth-param ]

2895	       realm          = "Realm" "=" quoted-string

2897	   The realm attribute is required for all access authentication
2898	   schemes which issue a challenge. The realm value, in combination
2899	   with the root URL of the server being accessed, defines the
2900	   authorization space. These realms allow the protected resources on
2901	   a server to be partitioned into a set of authorization spaces, each
2902	   with its own authentication scheme and/or database. The realm value
2903	   is a string, generally assigned by the origin server, which may
2904	   have additional semantics specific to the authentication scheme.

2906	   A user agent that wishes to authenticate itself with a server
2907	   (usually, but not necessarily, after receiving a 401 response), may
2908	   do so by including an Authorization header field with the request.
2909	   The Authorization field value consists of credentials containing
2910	   the authentication information of the user agent for the realm of
2911	   the resource being requested.

2913	       credentials    = auth-scheme [ 1*LWS encoded-cookie ] #auth-param

2915	       encoded-cookie = 1*<any CHAR except CTLs or tspecials,
2916	                           but including "=" and "/">

2918	   The domain over which credentials can be automatically applied by a
2919	   user agent is determined by the authorization space. If a request
2920	   is authenticated, the credentials can be reused for all other
2921	   requests within that authorization space for a period of time
2922	   determined by the authentication scheme, parameters, and/or user
2923	   preference.

2925	   The HTTP protocol does not restrict applications to this simple
2926	   challenge-response mechanism for access authentication. Additional
2927	   mechanisms may be used at the transport level, via message
2928	   encapsulation, and/or with additional header fields specifying
2929	   authentication information. However, these additional mechanisms
2930	   are not defined by this specification.

2932	   Proxies must be completely transparent regarding user agent access
2933	   authentication. That is, they must forward the WWW-Authenticate and
2934	   Authorization headers untouched. HTTP/1.0 does not provide a means
2935	   for a client to be authenticated with a proxy -- this feature will
2936	   be available in future versions of HTTP.

2938	       Note: The names Proxy-Authenticate and Proxy-Authorization
2939	       have been suggested as headers, analogous to WWW-Authenticate
2940	       and Authorization, but applying only to the immediate
2941	       connection with a proxy.

2943	10.1  Basic Authentication Scheme

2945	   The basic authentication scheme is based on the model that the
2946	   client must authenticate itself with a user-ID and a password for
2947	   each realm. The realm value should be considered an opaque string
2948	   which can only be compared for equality with other realms. The
2949	   server will service the request only if it can validate the user-ID
2950	   and password for the domain of the Request-URI.

2952	       basic-challenge   = "Basic" SP realm

2954	   The client sends the user-ID and password (separated by a single
2955	   colon ":" character) within a base64 [6] encoded-cookie in the
2956	   credentials.

2958	       basic-credentials = "Basic" SP basic-cookie
2959	       basic-cookie      = <base64 encoding of userid-password>
2960	       userid-password   = [ token ] ":" *text

2962	   There are no optional authentication parameters for the basic
2963	   scheme. For example, if the user agent wishes to send the user-ID
2964	   "Aladdin" and password "open sesame", it would use the following
2965	   header field:

2967	       Authorization: Basic QWxhZGRpbjpvcGVuIHNlc2FtZQ==

2969	   The basic authentication scheme is a non-secure method of filtering
2970	   unauthorized access to resources on an HTTP server. It is based on
2971	   the assumption that the connection between the client and the
2972	   server can be regarded as a trusted carrier. As this is not
2973	   generally true on an open network, the basic authentication scheme
2974	   should be used accordingly. In spite of this, clients are strongly
2975	   encouraged to implement the scheme in order to communicate with
2976	   servers that use it.

2978	11. Security Considerations

2980	   This section is meant to inform application developers, information
2981	   providers, and users of the security limitations in HTTP/1.0 as
2982	   described by this document. The discussion does not include
2983	   definitive solutions to the problems revealed, though it does make
2984	   some suggestions for reducing security risks.

2986	11.1  Authentication of Clients

2988	   As mentioned in Section 10.1, the Basic authentication scheme is
2989	   not considered to be a secure method of user authentication, nor
2990	   does it prevent the Entity-Body from being transmitted in clear
2991	   text across the physical network used as the carrier. HTTP/1.0 does
2992	   not prevent additional authentication schemes and encryption
2993	   mechanisms to be employed to increase security.

2995	11.2  Idempotent Methods

2997	   The writers of client software should be aware that the software
2998	   represents the user in their interactions over the net, and should
2999	   be careful to allow the user to be aware of any actions they may
3000	   take which may have an unexpected significance to themselves or
3001	   others.

3003	   In particular, the convention has been established that the GET and
3004	   HEAD methods should never have the significance of taking an
3005	   action. The link "click here to subscribe"--causing the reading of a
3006	   special "magic" document--is open to abuse by others making a link
3007	   "click here to see a pretty picture." These methods should be
3008	   considered "safe" and should not have side effects. This allows the
3009	   client software to represent other methods (such as POST, PUT and
3010	   DELETE) in a special way, so that the user is aware of the fact
3011	   that an action is being requested.

3013	11.3  Abuse of Server Log Information

3015	   A server is in the position to save personal data about a user's
3016	   requests which may identify their reading patterns or subjects of
3017	   interest. This information is clearly confidential in nature and
3018	   its handling may be constrained by law in certain countries. People
3019	   using the HTTP protocol to provide data are responsible for
3020	   ensuring that such material is not distributed without the
3021	   permission of any individuals that are identifiable by the
3022	   published results.

3024	   Two header fields are worth special mention in this context:
3025	   Referer and From. The Referer field allows reading patterns to be
3026	   studied and reverse links drawn. Although it can be very useful,
3027	   its power can be abused if user details are not separated from the
3028	   information contained in the Referer. Even when the personal
3029	   information has been removed, the Referer field may have indicated
3030	   a secure document's URI, whose revelation itself would be a breach
3031	   of security.

3033	   The information sent in the From field might conflict with the
3034	   user's privacy interests or their site's security policy, and hence
3035	   it should not be transmitted without the user being able to
3036	   disable, enable, and modify the contents of the field. The user
3037	   must be able to set the active contents of this field within a user
3038	   preference or application defaults configuration.

3040	   We suggest, though do not require, that a convenient toggle
3041	   interface be provided for the user to enable or disable the sending
3042	   of From and Referer information.

3044	12.  Acknowledgments

3046	   This specification makes heavy use of the augmented BNF and generic
3047	   constructs defined by David H. Crocker for RFC 822 [8]. Similarly,
3048	   it reuses many of the definitions provided by Nathaniel Borenstein
3049	   and Ned Freed for MIME [6]. We hope that their inclusion in this
3050	   specification will help reduce past confusion over the relationship
3051	   between HTTP/1.0 and Internet mail.

3053	   The HTTP protocol has evolved considerably over the past three
3054	   years. It has benefited from a large and active developer community--
3055	   the many people who have participated on the www-talk mailing list--
3056	   and it is that community which has been most responsible for the
3057	   success of HTTP and of the World-Wide Web in general.
3058	   Marc Andreessen, Robert Cailliau, Daniel W. Connolly, Bob Denny,
3059	   Phillip M. Hallam-Baker, Haringkon W. Lie, Ari Luotonen, Rob McCool,
3060	   Dave Raggett, Tony Sanders, and Marc VanHeyningen deserve special
3061	   recognition for their efforts in defining aspects of the protocol
3062	   for early versions of this specification.

3064	   This document has benefited greatly from the comments of all those
3065	   participating in the HTTPWG. In addition to those already
3066	   mentioned, the following individuals have contributed to this
3067	   specification:

3069	       Gary Adams                         Harald Tveit Alvestrand
3070	       Keith Ball                         Brian Behlendorf
3071	       Paul Burchard                      Maurizio Codogno
3072	       Mike Cowlishaw                     Michael A. Dolan
3073	       John Franks                        Alex Hopmann
3074	       Bob Jernigan                       Martijn Koster
3075	       Dave Kristol                       Daniel LaLiberte
3076	       Albert Lunde                       John C. Mallery
3077	       Larry Masinter                     Mitra
3078	       Gavin Nicol                        Marc Salomon
3079	       Chuck Shotton                      Eric W. Sink
3080	       Simon E. Spero

3082	13. References

3084	   [1]  H. T. Alvestrand. "Tags for the identification of languages."
3085	        RFC 1766, UNINETT, March 1995.

3087	   [2]  F. Anklesaria, M. McCahill, P. Lindner, D. Johnson, D. Torrey,
3088	        and B. Alberti. "The Internet Gopher Protocol: A distributed
3089	        document search and retrieval protocol." RFC 1436, University
3090	        of Minnesota, March 1993.

3092	   [3]  T. Berners-Lee. "Universal Resource Identifiers in WWW: A
3093	        Unifying Syntax for the Expression of Names and Addresses of
3094	        Objects on the Network as used in the World-Wide Web."
3095	        RFC 1630, CERN, June 1994.

3097	   [4]  T. Berners-Lee and D. Connolly. "HyperText Markup Language
3098	        Specification - 2.0." Work in Progress (draft-ietf-html-spec-
3099	        01.txt), CERN, HaL Computer Systems, February 1995.

3101	   [5]  T. Berners-Lee, L. Masinter, and M. McCahill. "Uniform Resource
3102	        Locators (URL)." RFC 1738, CERN, Xerox PARC, University of
3103	        Minnesota, October 1994.

3105	   [6]  N. Borenstein and N. Freed. "MIME (Multipurpose Internet Mail
3106	        Extensions) Part One: Mechanisms for Specifying and Describing
3107	        the Format of Internet Message Bodies." RFC 1521, Bellcore,
3108	        Innosoft, September 1993.

3110	   [7]  R. Braden. "Requirements for Internet hosts - application and
3111	        support." STD 3, RFC 1123, IETF, October 1989.

3113	   [8]  D. H. Crocker. "Standard for the Format of ARPA Internet Text
3114	        Messages." STD 11, RFC 822, UDEL, August 1982.

3116	   [9]  F. Davis, B. Kahle, H. Morris, J. Salem, T. Shen, R. Wang,
3117	        J. Sui, and M. Grinbaum. "WAIS Interface Protocol Prototype
3118	        Functional Specification." (v1.5), Thinking Machines
3119	        Corporation, April 1990.

3121	   [10] R. T. Fielding. "Relative Uniform Resource Locators." Work in
3122	        Progress (draft-ietf-uri-relative-url-05.txt), UC Irvine,
3123	        January 1995.

3125	   [11] M. Horton and R. Adams. "Standard for interchange of USENET
3126	        messages." RFC 1036 (Obsoletes RFC 850), AT&T Bell
3127	        Laboratories, Center for Seismic Studies, December 1987.

3129	   [12] B. Kantor and P. Lapsley. "Network News Transfer Protocol: A
3130	        Proposed Standard for the Stream-Based Transmission of News."
3131	        RFC 977, UC San Diego, UC Berkeley, February 1986.

3133	   [13] K. Moore. "MIME (Multipurpose Internet Mail Extensions) Part
3134	        Two: Message Header Extensions for Non-ASCII Text." RFC 1522,
3135	        University of Tennessee, September 1993.

3137	   [14] J. Postel. "Simple Mail Transfer Protocol." STD 10, RFC 821,
3138	        USC/ISI, August 1982.

3140	   [15] J. Postel. "Media Type Registration Procedure." RFC 1590,
3141	        USC/ISI, March 1994.

3143	   [16] J. Postel and J. K. Reynolds. "File Transfer Protocol (FTP)."
3144	        STD 9, RFC 959, USC/ISI, October 1985.

3146	   [17] J. Reynolds and J. Postel. "Assigned Numbers." STD 2, RFC 1700,
3147	        USC/ISI, October 1994.

3149	   [18] US-ASCII. Coded Character Set - 7-Bit American Standard Code for
3150	        Information Interchange. Standard ANSI X3.4-1986, ANSI, 1986.

3152	   [19] ISO-8859. International Standard -- Information Processing --
3153	        8-bit Single-Byte Coded Graphic Character Sets -- Part 1: Latin
3154	        Alphabet No. 1, ISO 8859-1:1987. Part 2: Latin alphabet No. 2,
3155	        ISO 8859-2, 1987. Part 3: Latin alphabet No. 3, ISO 8859-3,
3156	        1988. Part 4: Latin alphabet No. 4, ISO 8859-4, 1988. Part 5:
3157	        Latin/Cyrillic alphabet, ISO 8859-5, 1988. Part 6: Latin/Arabic
3158	        alphabet, ISO 8859-6, 1987. Part 7: Latin/Greek alphabet, ISO
3159	        8859-7, 1987. Part 8: Latin/Hebrew alphabet, ISO 8859-8, 1988.
3160	        Part 9: Latin alphabet No. 5, ISO 8859-9, 1990.

3162	14.  Authors' Addresses

3164	   Tim Berners-Lee
3165	   Director, W3 Consortium
3166	   MIT Laboratory for Computer Science
3167	   545 Technology Square
3168	   Cambridge, MA 02139, U.S.A.
3169	   Tel: +1 (617) 253 9670
3170	   Fax: +1 (617) 258 8682
3171	   Email: timbl@w3.org

3173	   Roy T. Fielding
3174	   Department of Information and Computer Science
3175	   University of California
3176	   Irvine, CA 92717-3425, U.S.A.
3177	   Tel: +1 (714) 824-4049
3178	   Fax: +1 (714) 824-4056
3179	   Email: fielding@ics.uci.edu

3181	   Henrik Frystyk Nielsen
3182	   World-Wide Web Project
3183	   CERN,
3184	   1211 Geneva 23, Switzerland
3185	   Tel: +41 (22) 767 8265
3186	   Fax: +41 (22) 767 8730
3187	   Email: frystyk@w3.org

3189	Appendices

3191	   These appendices are provided for informational reasons only -- they
3192	   do not form a part of the HTTP/1.0 specification.

3194	A.  Internet Media Type message/http

3196	   In addition to defining the HTTP/1.0 protocol, this document serves
3197	   as the specification for the Internet media type "message/http".
3198	   The following is to be registered with IANA [15].

3200	       Media Type name:         message

3202	       Media subtype name:      http

3204	       Required parameters:     none

3206	       Optional parameters:     version, type

3208	             version: The HTTP-Version number of the enclosed message
3209	                      (e.g. "1.0"). If not present, the version can be
3210	                      determined from the first line of the body.

3212	             type:    The message type -- "request" or "response". If
3213	                      not present, the type can be determined from the
3214	                      first line of the body.

3216	       Encoding considerations: only "7bit", "8bit", or "binary" are
3217	                                permitted

3219	       Security considerations: none

3221	B.  Minimum Compliance

3223	   Early reviews of this specification have indicated the need for a
3224	   statement of the minimum requirements for an implementation to be
3225	   considered in compliance with HTTP/1.0. This section will be
3226	   written soon.

3228	       Note: The primary difficulty in determining a standard for
3229	       minimum compliance rests in the fact that HTTP is a flexible
3230	       protocol which can be used for many purposes. The
3231	       requirements for special purpose applications often differ
3232	       from those of general purpose applications.

3234	B.1  Servers

3236	   Servers have a special responsibility for being honest when
3237	   generating their responses to requesting clients. The Status-Code
3238	   sent in the Status-Line must correspond to the actual action taken
3239	   by the server. This is especially the case when the method used in
3240	   the request is one of PUT, POST, DELETE, LINK, and UNLINK. If a
3241	   Status-Code of 200 is returned, the client must be able to assume
3242	   that the action has been carried out. If the server is not able to
3243	   fulfill the requested action immediately, the correct status code
3244	   to use is "202 Accepted".

3246	   The methods GET and HEAD must be supported by all general-purpose
3247	   servers. Servers which provide Last-Modified dates for resources
3248	   must also support the conditional GET method.

3250	C.  Tolerant Applications

3252	   While it may be appropriate for testing applications to verify full
3253	   conformance to this specification, it is recommended that
3254	   operational applications be tolerant of deviations. This appendix
3255	   mentions the most important topics where tolerance is recommended.

3257	C.1  Request-Line, Status-Line, and Header Fields

3259	   Clients should be tolerant in parsing the Status-Line and servers
3260	   tolerant when parsing the Request-Line. In particular, they should
3261	   accept any amount of SP and HTAB characters between fields, even
3262	   though only a single SP is specified.

3264	   The line terminator for HTTP-header fields should be the sequence
3265	   CRLF. However, we recommend that applications, when parsing such
3266	   headers, recognize a single LF as a line terminator and ignore the
3267	   leading CR.

3269	   We recommend that servers allocate URIs free of "variant"
3270	   characters (characters whose representation differs in some of the
3271	   national variant character sets), punctuation characters, and
3272	   spaces. This makes URIs easier to handle by humans when the need
3273	   arises (such as for debugging or transmission through non hypertext
3274	   systems).

3276	D.  Relationship to MIME

3278	   HTTP/1.0 reuses many of the constructs defined for Internet Mail
3279	   (RFC 822 [8]) and the Multipurpose Internet Mail Extensions
3280	   (MIME [6]) to allow entities to be transmitted in an open variety
3281	   of representations and with extensible mechanisms. However, HTTP is
3282	   not a MIME-conforming application. HTTP's performance requirements
3283	   differ substantially from those of Internet mail. Since it is not
3284	   limited by the restrictions of existing mail protocols and
3285	   gateways, HTTP does not obey some of the constraints imposed by
3286	   RFC 822 and MIME for mail transport.

3288	   This appendix describes specific areas where HTTP differs from
3289	   MIME. Gateways to MIME-compliant protocols must be aware of these
3290	   differences and provide the appropriate conversions where
3291	   necessary. No conversion should be necessary for a MIME-conforming
3292	   entity to be transferred using HTTP.

3294	D.1  Conversion to Canonical Form

3296	   MIME requires that an entity be converted to canonical form prior
3297	   to being transferred, as described in Appendix G of RFC 1521 [6].
3298	   Although HTTP does require media types to be transferred in
3299	   canonical form, it changes the definition of "canonical form" for
3300	   text-based media types as described in Section 8.1.1.

3302	D.1.1 Representation of Line Breaks

3304	   MIME requires that the canonical form of any text type represent
3305	   line breaks as CRLF and forbids the use of CR or LF outside of line
3306	   break sequences. Since HTTP allows CRLF, bare CR, and bare LF (or
3307	   the octet sequence(s) to which they would be translated for the
3308	   given character set) to indicate a line break within text content,
3309	   recipients of an HTTP message cannot rely upon receiving MIME-
3310	   canonical line breaks in text.

3312	   Where it is possible, a gateway from HTTP to a MIME-conformant
3313	   protocol should translate all line breaks within text/* media types
3314	   to the MIME canonical form of CRLF. However, this may be
3315	   complicated by the presence of a Content-Encoding and by the fact
3316	   that HTTP allows the use of some character sets which do not use
3317	   octets 13 and 10 to represent CR and LF (as is the case for some
3318	   multi-byte character sets).

3320	D.1.2 Default Character Set

3322	   MIME requires that all subtypes of the top-level Content-Type
3323	   "text" have a default character set of US-ASCII [18]. In contrast,
3324	   HTTP defines the default character set for "text" to be
3325	   ISO88591 [19] (a superset of US-ASCII). Therefore, if a text/*
3326	   media type given in the Content-Type header field does not already
3327	   include an explicit charset parameter, the parameter

3329	       ;charset="iso-8859-1"

3331	   should be added by the gateway if the entity contains any octets
3332	   greater than 127.

3334	D.2  Default Content-Transfer-Encoding

3336	   The default Content-Transfer-Encoding (CTE) for all MIME messages
3337	   is "7bit". In contrast, HTTP defines the default CTE to be
3338	   "binary". Therefore, if an entity does not include an explicit CTE
3339	   header field, the gateway should apply either the "quoted-printable"
3340	   or "base64" transfer encodings and add the appropriate
3341	   Content-Transfer-Encoding field. At a minimum, the explicit CTE
3342	   field of

3344	       Content-Transfer-Encoding: binary

3346	   should be added by the gateway if it is unwilling to apply a mail-
3347	   safe encoding.

3349	D.3  Introduction of Content-Encoding

3351	   MIME does not include any concept equivalent to HTTP's Content-
3352	   Encoding header field. Since this acts as a modifier on the media
3353	   type, gateways to MIME-conformant protocols should either change
3354	   the value of the Content-Type header field or decode the Entity-
3355	   Body before forwarding the message.

3357	       Note: Some experimental applications of Content-Type for
3358	       Internet mail have used a media-type parameter of
3359	       ";conversions=<encoding-mechanisms>" to perform an
3360	       equivalent function as Content-Encoding. However, this
3361	       parameter is not part of the MIME specification at the time
3362	       of this writing.

3364	E.  Example of Version Control

3366	   This appendix gives an example on how the Entity-Header fields
3367	   Version and Derived-From can be used to apply version control to
3368	   the creation and parallel development of a work. In order to
3369	   simplify the example, only two user agents (A and B) are considered
3370	   together with an origin server S.

3372	      o A sends a POST request to S, including the header
3373	        "Version: 1.0" and an entity

3375	      o S replies "201 Created" to A, including the header
3376	        "Version: 1.0" and a URI-header which should be used for
3377	        future references

3379	      o A starts editing the entity

3381	      o B sends a GET request to S

3383	      o S replies "200 OK" to B, including the entity with a header
3384	        "Version: 1.0"

3386	      o B starts editing the entity

3388	      o B sends a PUT request to S, including the entity and a header
3389	        "Derived-From: 1.0"

3391	      o S replies "204 No Content" to B, including a header
3392	        "Version: 1.1" but no entity

3394	      o A sends a PUT request to S, including the entity and a header
3395	        "Derived-From: 1.0"

3397	      o S replies "409 Conflict" to A, including "Version: 1.1" and the
3398	        list of problems with merging A's changes to 1.0 with those
3399	        already applied for B and version 1.1

3401	      o A merges B's changes with its own, possibly with help from the
3402	        user of A

3404	      o A sends a PUT request to S including the entity and the header
3405	        "Derived-From: 1.1"

3407	      o S replies "204 No Content" to A, including the header
3408	        "Version: 1.2" but no entity

3410	   The example can be expanded to any number of involved user agents,
3411	   though the likelihood of conflicts and the difficulty of resolving
3412	   them may increase.