idnits 2.17.00 (12 Aug 2021) /tmp/idnits14625/draft-ietf-httpbis-messaging-13.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- == There is 1 instance of lines with non-ascii characters in the document. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- -- The draft header indicates that this document obsoletes RFC7230, but the abstract doesn't seem to directly say this. It does mention RFC7230 though, so this could be OK. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year == Line 1795 has weird spacing: '...ncoding stan...' == The document seems to contain a disclaimer for pre-RFC5378 work, but was first submitted on or after 10 November 2008. The disclaimer is usually necessary only for documents that revise or obsolete older RFCs, and that take significant amounts of text from those RFCs. If you can contact all authors of the source material and they are willing to grant the BCP78 rights to the IETF Trust, you can and should remove the disclaimer. Otherwise, the disclaimer is needed and you can ignore this comment. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (December 14, 2020) is 522 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Unused Reference: 'RFC3986' is defined on line 1884, but no explicit reference was found in the text == Unused Reference: 'RFC7231' is defined on line 1974, but no explicit reference was found in the text == Outdated reference: A later version (-19) exists of draft-ietf-httpbis-cache-13 -- Possible downref: Normative reference to a draft: ref. 'Caching' ** Downref: Normative reference to an Informational RFC: RFC 1950 ** Downref: Normative reference to an Informational RFC: RFC 1951 ** Downref: Normative reference to an Informational RFC: RFC 1952 == Outdated reference: A later version (-19) exists of draft-ietf-httpbis-semantics-13 -- Possible downref: Normative reference to a draft: ref. 'Semantics' -- Possible downref: Non-RFC (?) normative reference: ref. 'USASCII' -- Possible downref: Non-RFC (?) normative reference: ref. 'Welch' -- Obsolete informational reference (is this intentional?): RFC 2068 (Obsoleted by RFC 2616) -- Duplicate reference: RFC7230, mentioned in 'RFC7230', was also mentioned in 'Err4667'. Summary: 3 errors (**), 0 flaws (~~), 8 warnings (==), 8 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 HTTP Working Group R. Fielding, Ed. 3 Internet-Draft Adobe 4 Obsoletes: 7230 (if approved) M. Nottingham, Ed. 5 Intended status: Standards Track Fastly 6 Expires: June 17, 2021 J. Reschke, Ed. 7 greenbytes 8 December 14, 2020 10 HTTP/1.1 11 draft-ietf-httpbis-messaging-13 13 Abstract 15 The Hypertext Transfer Protocol (HTTP) is a stateless application- 16 level protocol for distributed, collaborative, hypertext information 17 systems. This document specifies the HTTP/1.1 message syntax, 18 message parsing, connection management, and related security 19 concerns. 21 This document obsoletes portions of RFC 7230. 23 Editorial Note 25 This note is to be removed before publishing as an RFC. 27 Discussion of this draft takes place on the HTTP working group 28 mailing list (ietf-http-wg@w3.org), which is archived at 29 . 31 Working Group information can be found at ; 32 source code and issues list for this draft can be found at 33 . 35 The changes in this draft are summarized in Appendix D.14. 37 Status of This Memo 39 This Internet-Draft is submitted in full conformance with the 40 provisions of BCP 78 and BCP 79. 42 Internet-Drafts are working documents of the Internet Engineering 43 Task Force (IETF). Note that other groups may also distribute 44 working documents as Internet-Drafts. The list of current Internet- 45 Drafts is at https://datatracker.ietf.org/drafts/current/. 47 Internet-Drafts are draft documents valid for a maximum of six months 48 and may be updated, replaced, or obsoleted by other documents at any 49 time. It is inappropriate to use Internet-Drafts as reference 50 material or to cite them other than as "work in progress." 52 This Internet-Draft will expire on June 17, 2021. 54 Copyright Notice 56 Copyright (c) 2020 IETF Trust and the persons identified as the 57 document authors. All rights reserved. 59 This document is subject to BCP 78 and the IETF Trust's Legal 60 Provisions Relating to IETF Documents (https://trustee.ietf.org/ 61 license-info) in effect on the date of publication of this document. 62 Please review these documents carefully, as they describe your rights 63 and restrictions with respect to this document. Code Components 64 extracted from this document must include Simplified BSD License text 65 as described in Section 4.e of the Trust Legal Provisions and are 66 provided without warranty as described in the Simplified BSD License. 68 This document may contain material from IETF Documents or IETF 69 Contributions published or made publicly available before November 70 10, 2008. The person(s) controlling the copyright in some of this 71 material may not have granted the IETF Trust the right to allow 72 modifications of such material outside the IETF Standards Process. 73 Without obtaining an adequate license from the person(s) controlling 74 the copyright in such materials, this document may not be modified 75 outside the IETF Standards Process, and derivative works of it may 76 not be created outside the IETF Standards Process, except to format 77 it for publication as an RFC or to translate it into languages other 78 than English. 80 Table of Contents 82 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 4 83 1.1. Requirements Notation . . . . . . . . . . . . . . . . . . 5 84 1.2. Syntax Notation . . . . . . . . . . . . . . . . . . . . . 5 85 2. Message . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 86 2.1. Message Format . . . . . . . . . . . . . . . . . . . . . 6 87 2.2. Message Parsing . . . . . . . . . . . . . . . . . . . . . 7 88 2.3. HTTP Version . . . . . . . . . . . . . . . . . . . . . . 8 89 3. Request Line . . . . . . . . . . . . . . . . . . . . . . . . 9 90 3.1. Method . . . . . . . . . . . . . . . . . . . . . . . . . 10 91 3.2. Request Target . . . . . . . . . . . . . . . . . . . . . 10 92 3.2.1. origin-form . . . . . . . . . . . . . . . . . . . . . 11 93 3.2.2. absolute-form . . . . . . . . . . . . . . . . . . . . 11 94 3.2.3. authority-form . . . . . . . . . . . . . . . . . . . 12 95 3.2.4. asterisk-form . . . . . . . . . . . . . . . . . . . . 12 96 3.3. Reconstructing the Target URI . . . . . . . . . . . . . . 13 97 4. Status Line . . . . . . . . . . . . . . . . . . . . . . . . . 14 98 5. Field Syntax . . . . . . . . . . . . . . . . . . . . . . . . 15 99 5.1. Field Line Parsing . . . . . . . . . . . . . . . . . . . 16 100 5.2. Obsolete Line Folding . . . . . . . . . . . . . . . . . . 16 101 6. Message Body . . . . . . . . . . . . . . . . . . . . . . . . 17 102 6.1. Transfer-Encoding . . . . . . . . . . . . . . . . . . . . 17 103 6.2. Content-Length . . . . . . . . . . . . . . . . . . . . . 19 104 6.3. Message Body Length . . . . . . . . . . . . . . . . . . . 19 105 7. Transfer Codings . . . . . . . . . . . . . . . . . . . . . . 22 106 7.1. Chunked Transfer Coding . . . . . . . . . . . . . . . . . 22 107 7.1.1. Chunk Extensions . . . . . . . . . . . . . . . . . . 23 108 7.1.2. Chunked Trailer Section . . . . . . . . . . . . . . . 23 109 7.1.3. Decoding Chunked . . . . . . . . . . . . . . . . . . 24 110 7.2. Transfer Codings for Compression . . . . . . . . . . . . 24 111 7.3. Transfer Coding Registry . . . . . . . . . . . . . . . . 25 112 7.4. Negotiating Transfer Codings . . . . . . . . . . . . . . 25 113 8. Handling Incomplete Messages . . . . . . . . . . . . . . . . 26 114 9. Connection Management . . . . . . . . . . . . . . . . . . . . 27 115 9.1. Establishment . . . . . . . . . . . . . . . . . . . . . . 27 116 9.2. Associating a Response to a Request . . . . . . . . . . . 28 117 9.3. Persistence . . . . . . . . . . . . . . . . . . . . . . . 28 118 9.3.1. Retrying Requests . . . . . . . . . . . . . . . . . . 29 119 9.3.2. Pipelining . . . . . . . . . . . . . . . . . . . . . 29 120 9.4. Concurrency . . . . . . . . . . . . . . . . . . . . . . . 30 121 9.5. Failures and Timeouts . . . . . . . . . . . . . . . . . . 31 122 9.6. Tear-down . . . . . . . . . . . . . . . . . . . . . . . . 31 123 9.7. TLS Connection Initiation . . . . . . . . . . . . . . . . 33 124 9.8. TLS Connection Closure . . . . . . . . . . . . . . . . . 33 125 10. Enclosing Messages as Data . . . . . . . . . . . . . . . . . 34 126 10.1. Media Type message/http . . . . . . . . . . . . . . . . 34 127 10.2. Media Type application/http . . . . . . . . . . . . . . 35 128 11. Security Considerations . . . . . . . . . . . . . . . . . . . 36 129 11.1. Response Splitting . . . . . . . . . . . . . . . . . . . 36 130 11.2. Request Smuggling . . . . . . . . . . . . . . . . . . . 37 131 11.3. Message Integrity . . . . . . . . . . . . . . . . . . . 38 132 11.4. Message Confidentiality . . . . . . . . . . . . . . . . 38 133 12. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 38 134 12.1. Field Name Registration . . . . . . . . . . . . . . . . 39 135 12.2. Media Type Registration . . . . . . . . . . . . . . . . 39 136 12.3. Transfer Coding Registration . . . . . . . . . . . . . . 39 137 12.4. ALPN Protocol ID Registration . . . . . . . . . . . . . 40 138 13. References . . . . . . . . . . . . . . . . . . . . . . . . . 40 139 13.1. Normative References . . . . . . . . . . . . . . . . . . 40 140 13.2. Informative References . . . . . . . . . . . . . . . . . 42 141 Appendix A. Collected ABNF . . . . . . . . . . . . . . . . . . . 43 142 Appendix B. Differences between HTTP and MIME . . . . . . . . . 45 143 B.1. MIME-Version . . . . . . . . . . . . . . . . . . . . . . 45 144 B.2. Conversion to Canonical Form . . . . . . . . . . . . . . 45 145 B.3. Conversion of Date Formats . . . . . . . . . . . . . . . 46 146 B.4. Conversion of Content-Encoding . . . . . . . . . . . . . 46 147 B.5. Conversion of Content-Transfer-Encoding . . . . . . . . . 46 148 B.6. MHTML and Line Length Limitations . . . . . . . . . . . . 46 149 Appendix C. Changes from previous RFCs . . . . . . . . . . . . . 47 150 C.1. Changes from HTTP/0.9 . . . . . . . . . . . . . . . . . . 47 151 C.2. Changes from HTTP/1.0 . . . . . . . . . . . . . . . . . . 47 152 C.2.1. Multihomed Web Servers . . . . . . . . . . . . . . . 47 153 C.2.2. Keep-Alive Connections . . . . . . . . . . . . . . . 47 154 C.2.3. Introduction of Transfer-Encoding . . . . . . . . . . 48 155 C.3. Changes from RFC 7230 . . . . . . . . . . . . . . . . . . 48 156 Appendix D. Change Log . . . . . . . . . . . . . . . . . . . . . 49 157 D.1. Between RFC7230 and draft 00 . . . . . . . . . . . . . . 49 158 D.2. Since draft-ietf-httpbis-messaging-00 . . . . . . . . . . 49 159 D.3. Since draft-ietf-httpbis-messaging-01 . . . . . . . . . . 49 160 D.4. Since draft-ietf-httpbis-messaging-02 . . . . . . . . . . 50 161 D.5. Since draft-ietf-httpbis-messaging-03 . . . . . . . . . . 50 162 D.6. Since draft-ietf-httpbis-messaging-04 . . . . . . . . . . 51 163 D.7. Since draft-ietf-httpbis-messaging-05 . . . . . . . . . . 51 164 D.8. Since draft-ietf-httpbis-messaging-06 . . . . . . . . . . 52 165 D.9. Since draft-ietf-httpbis-messaging-07 . . . . . . . . . . 52 166 D.10. Since draft-ietf-httpbis-messaging-08 . . . . . . . . . . 52 167 D.11. Since draft-ietf-httpbis-messaging-09 . . . . . . . . . . 52 168 D.12. Since draft-ietf-httpbis-messaging-10 . . . . . . . . . . 52 169 D.13. Since draft-ietf-httpbis-messaging-11 . . . . . . . . . . 53 170 D.14. Since draft-ietf-httpbis-messaging-12 . . . . . . . . . . 53 171 Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . 54 172 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 54 174 1. Introduction 176 The Hypertext Transfer Protocol (HTTP) is a stateless application- 177 level request/response protocol that uses extensible semantics and 178 self-descriptive messages for flexible interaction with network-based 179 hypertext information systems. HTTP/1.1 is defined by: 181 o This document 183 o "HTTP Semantics" [Semantics] 185 o "HTTP Caching" [Caching] 187 This document specifies how HTTP semantics are conveyed using the 188 HTTP/1.1 message syntax, framing and connection management 189 mechanisms. Its goal is to define the complete set of requirements 190 for HTTP/1.1 message parsers and message-forwarding intermediaries. 192 This document obsoletes the portions of RFC 7230 related to HTTP/1.1 193 messaging and connection management, with the changes being 194 summarized in Appendix C.3. The other parts of RFC 7230 are 195 obsoleted by "HTTP Semantics" [Semantics]. 197 1.1. Requirements Notation 199 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 200 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 201 "OPTIONAL" in this document are to be interpreted as described in BCP 202 14 [RFC2119] [RFC8174] when, and only when, they appear in all 203 capitals, as shown here. 205 Conformance criteria and considerations regarding error handling are 206 defined in Section 2 of [Semantics]. 208 1.2. Syntax Notation 210 This specification uses the Augmented Backus-Naur Form (ABNF) 211 notation of [RFC5234], extended with the notation for case- 212 sensitivity in strings defined in [RFC7405]. 214 It also uses a list extension, defined in Section 5.6.1 of 215 [Semantics], that allows for compact definition of comma-separated 216 lists using a '#' operator (similar to how the '*' operator indicates 217 repetition). Appendix A shows the collected grammar with all list 218 operators expanded to standard ABNF notation. 220 As a convention, ABNF rule names prefixed with "obs-" denote 221 "obsolete" grammar rules that appear for historical reasons. 223 The following core rules are included by reference, as defined in 224 [RFC5234], Appendix B.1: ALPHA (letters), CR (carriage return), CRLF 225 (CR LF), CTL (controls), DIGIT (decimal 0-9), DQUOTE (double quote), 226 HEXDIG (hexadecimal 0-9/A-F/a-f), HTAB (horizontal tab), LF (line 227 feed), OCTET (any 8-bit sequence of data), SP (space), and VCHAR (any 228 visible [USASCII] character). 230 The rules below are defined in [Semantics]: 232 BWS = 233 OWS = 234 RWS = 235 absolute-URI = 236 absolute-path = 237 authority = 238 comment = 239 field-name = 240 field-value = 241 obs-text = 242 port = 243 query = 244 quoted-string = 245 token = 246 transfer-coding = 247 248 uri-host = 250 2. Message 252 2.1. Message Format 254 An HTTP/1.1 message consists of a start-line followed by a CRLF and a 255 sequence of octets in a format similar to the Internet Message Format 256 [RFC5322]: zero or more header field lines (collectively referred to 257 as the "headers" or the "header section"), an empty line indicating 258 the end of the header section, and an optional message body. 260 HTTP-message = start-line CRLF 261 *( field-line CRLF ) 262 CRLF 263 [ message-body ] 265 A message can be either a request from client to server or a response 266 from server to client. Syntactically, the two types of message 267 differ only in the start-line, which is either a request-line (for 268 requests) or a status-line (for responses), and in the algorithm for 269 determining the length of the message body (Section 6). 271 start-line = request-line / status-line 273 In theory, a client could receive requests and a server could receive 274 responses, distinguishing them by their different start-line formats. 275 In practice, servers are implemented to only expect a request (a 276 response is interpreted as an unknown or invalid request method) and 277 clients are implemented to only expect a response. 279 Although HTTP makes use of some protocol elements similar to the 280 Multipurpose Internet Mail Extensions (MIME) [RFC2045], see 281 Appendix B for the differences between HTTP and MIME messages. 283 2.2. Message Parsing 285 The normal procedure for parsing an HTTP message is to read the 286 start-line into a structure, read each header field line into a hash 287 table by field name until the empty line, and then use the parsed 288 data to determine if a message body is expected. If a message body 289 has been indicated, then it is read as a stream until an amount of 290 octets equal to the message body length is read or the connection is 291 closed. 293 A recipient MUST parse an HTTP message as a sequence of octets in an 294 encoding that is a superset of US-ASCII [USASCII]. Parsing an HTTP 295 message as a stream of Unicode characters, without regard for the 296 specific encoding, creates security vulnerabilities due to the 297 varying ways that string processing libraries handle invalid 298 multibyte character sequences that contain the octet LF (%x0A). 299 String-based parsers can only be safely used within protocol elements 300 after the element has been extracted from the message, such as within 301 a header field line value after message parsing has delineated the 302 individual field lines. 304 Although the line terminator for the start-line and fields is the 305 sequence CRLF, a recipient MAY recognize a single LF as a line 306 terminator and ignore any preceding CR. 308 A sender MUST NOT generate a bare CR (a CR character not immediately 309 followed by LF) within any protocol elements other than the payload 310 data. A recipient of such a bare CR MUST consider that element to be 311 invalid or replace each bare CR with SP before processing the element 312 or forwarding the message. 314 Older HTTP/1.0 user agent implementations might send an extra CRLF 315 after a POST request as a workaround for some early server 316 applications that failed to read message body content that was not 317 terminated by a line-ending. An HTTP/1.1 user agent MUST NOT preface 318 or follow a request with an extra CRLF. If terminating the request 319 message body with a line-ending is desired, then the user agent MUST 320 count the terminating CRLF octets as part of the message body length. 322 In the interest of robustness, a server that is expecting to receive 323 and parse a request-line SHOULD ignore at least one empty line (CRLF) 324 received prior to the request-line. 326 A sender MUST NOT send whitespace between the start-line and the 327 first header field. A recipient that receives whitespace between the 328 start-line and the first header field MUST either reject the message 329 as invalid or consume each whitespace-preceded line without further 330 processing of it (i.e., ignore the entire line, along with any 331 subsequent lines preceded by whitespace, until a properly formed 332 header field is received or the header section is terminated). 334 The presence of such whitespace in a request might be an attempt to 335 trick a server into ignoring that field line or processing the line 336 after it as a new request, either of which might result in a security 337 vulnerability if other implementations within the request chain 338 interpret the same message differently. Likewise, the presence of 339 such whitespace in a response might be ignored by some clients or 340 cause others to cease parsing. 342 When a server listening only for HTTP request messages, or processing 343 what appears from the start-line to be an HTTP request message, 344 receives a sequence of octets that does not match the HTTP-message 345 grammar aside from the robustness exceptions listed above, the server 346 SHOULD respond with a 400 (Bad Request) response. 348 2.3. HTTP Version 350 HTTP uses a "." numbering scheme to indicate versions 351 of the protocol. This specification defines version "1.1". 352 Section 2.5 of [Semantics] specifies the semantics of HTTP version 353 numbers. 355 The version of an HTTP/1.x message is indicated by an HTTP-version 356 field in the start-line. HTTP-version is case-sensitive. 358 HTTP-version = HTTP-name "/" DIGIT "." DIGIT 359 HTTP-name = %s"HTTP" 361 When an HTTP/1.1 message is sent to an HTTP/1.0 recipient [RFC1945] 362 or a recipient whose version is unknown, the HTTP/1.1 message is 363 constructed such that it can be interpreted as a valid HTTP/1.0 364 message if all of the newer features are ignored. This specification 365 places recipient-version requirements on some new features so that a 366 conformant sender will only use compatible features until it has 367 determined, through configuration or the receipt of a message, that 368 the recipient supports HTTP/1.1. 370 Intermediaries that process HTTP messages (i.e., all intermediaries 371 other than those acting as tunnels) MUST send their own HTTP-version 372 in forwarded messages. In other words, they are not allowed to 373 blindly forward the start-line without ensuring that the protocol 374 version in that message matches a version to which that intermediary 375 is conformant for both the receiving and sending of messages. 376 Forwarding an HTTP message without rewriting the HTTP-version might 377 result in communication errors when downstream recipients use the 378 message sender's version to determine what features are safe to use 379 for later communication with that sender. 381 A server MAY send an HTTP/1.0 response to an HTTP/1.1 request if it 382 is known or suspected that the client incorrectly implements the HTTP 383 specification and is incapable of correctly processing later version 384 responses, such as when a client fails to parse the version number 385 correctly or when an intermediary is known to blindly forward the 386 HTTP-version even when it doesn't conform to the given minor version 387 of the protocol. Such protocol downgrades SHOULD NOT be performed 388 unless triggered by specific client attributes, such as when one or 389 more of the request header fields (e.g., User-Agent) uniquely match 390 the values sent by a client known to be in error. 392 3. Request Line 394 A request-line begins with a method token, followed by a single space 395 (SP), the request-target, another single space (SP), and ends with 396 the protocol version. 398 request-line = method SP request-target SP HTTP-version 400 Although the request-line grammar rule requires that each of the 401 component elements be separated by a single SP octet, recipients MAY 402 instead parse on whitespace-delimited word boundaries and, aside from 403 the CRLF terminator, treat any form of whitespace as the SP separator 404 while ignoring preceding or trailing whitespace; such whitespace 405 includes one or more of the following octets: SP, HTAB, VT (%x0B), FF 406 (%x0C), or bare CR. However, lenient parsing can result in request 407 smuggling security vulnerabilities if there are multiple recipients 408 of the message and each has its own unique interpretation of 409 robustness (see Section 11.2). 411 HTTP does not place a predefined limit on the length of a request- 412 line, as described in Section 2 of [Semantics]. A server that 413 receives a method longer than any that it implements SHOULD respond 414 with a 501 (Not Implemented) status code. A server that receives a 415 request-target longer than any URI it wishes to parse MUST respond 416 with a 414 (URI Too Long) status code (see Section 15.5.15 of 417 [Semantics]). 419 Various ad hoc limitations on request-line length are found in 420 practice. It is RECOMMENDED that all HTTP senders and recipients 421 support, at a minimum, request-line lengths of 8000 octets. 423 3.1. Method 425 The method token indicates the request method to be performed on the 426 target resource. The request method is case-sensitive. 428 method = token 430 The request methods defined by this specification can be found in 431 Section 9 of [Semantics], along with information regarding the HTTP 432 method registry and considerations for defining new methods. 434 3.2. Request Target 436 The request-target identifies the target resource upon which to apply 437 the request. The client derives a request-target from its desired 438 target URI. There are four distinct formats for the request-target, 439 depending on both the method being requested and whether the request 440 is to a proxy. 442 request-target = origin-form 443 / absolute-form 444 / authority-form 445 / asterisk-form 447 No whitespace is allowed in the request-target. Unfortunately, some 448 user agents fail to properly encode or exclude whitespace found in 449 hypertext references, resulting in those disallowed characters being 450 sent as the request-target in a malformed request-line. 452 Recipients of an invalid request-line SHOULD respond with either a 453 400 (Bad Request) error or a 301 (Moved Permanently) redirect with 454 the request-target properly encoded. A recipient SHOULD NOT attempt 455 to autocorrect and then process the request without a redirect, since 456 the invalid request-line might be deliberately crafted to bypass 457 security filters along the request chain. 459 A client MUST send a Host header field in all HTTP/1.1 request 460 messages. If the target URI includes an authority component, then a 461 client MUST send a field value for Host that is identical to that 462 authority component, excluding any userinfo subcomponent and its "@" 463 delimiter (Section 4.2.1 of [Semantics]). If the authority component 464 is missing or undefined for the target URI, then a client MUST send a 465 Host header field with an empty field value. 467 A server MUST respond with a 400 (Bad Request) status code to any 468 HTTP/1.1 request message that lacks a Host header field and to any 469 request message that contains more than one Host header field line or 470 a Host header field with an invalid field value. 472 3.2.1. origin-form 474 The most common form of request-target is the _origin-form_. 476 origin-form = absolute-path [ "?" query ] 478 When making a request directly to an origin server, other than a 479 CONNECT or server-wide OPTIONS request (as detailed below), a client 480 MUST send only the absolute path and query components of the target 481 URI as the request-target. If the target URI's path component is 482 empty, the client MUST send "/" as the path within the origin-form of 483 request-target. A Host header field is also sent, as defined in 484 Section 7.2 of [Semantics]. 486 For example, a client wishing to retrieve a representation of the 487 resource identified as 489 http://www.example.org/where?q=now 491 directly from the origin server would open (or reuse) a TCP 492 connection to port 80 of the host "www.example.org" and send the 493 lines: 495 GET /where?q=now HTTP/1.1 496 Host: www.example.org 498 followed by the remainder of the request message. 500 3.2.2. absolute-form 502 When making a request to a proxy, other than a CONNECT or server-wide 503 OPTIONS request (as detailed below), a client MUST send the target 504 URI in _absolute-form_ as the request-target. 506 absolute-form = absolute-URI 508 The proxy is requested to either service that request from a valid 509 cache, if possible, or make the same request on the client's behalf 510 to either the next inbound proxy server or directly to the origin 511 server indicated by the request-target. Requirements on such 512 "forwarding" of messages are defined in Section 7.6 of [Semantics]. 514 An example absolute-form of request-line would be: 516 GET http://www.example.org/pub/WWW/TheProject.html HTTP/1.1 518 A client MUST send a Host header field in an HTTP/1.1 request even if 519 the request-target is in the absolute-form, since this allows the 520 Host information to be forwarded through ancient HTTP/1.0 proxies 521 that might not have implemented Host. 523 When a proxy receives a request with an absolute-form of request- 524 target, the proxy MUST ignore the received Host header field (if any) 525 and instead replace it with the host information of the request- 526 target. A proxy that forwards such a request MUST generate a new 527 Host field value based on the received request-target rather than 528 forward the received Host field value. 530 When an origin server receives a request with an absolute-form of 531 request-target, the origin server MUST ignore the received Host 532 header field (if any) and instead use the host information of the 533 request-target. Note that if the request-target does not have an 534 authority component, an empty Host header field will be sent in this 535 case. 537 To allow for transition to the absolute-form for all requests in some 538 future version of HTTP, a server MUST accept the absolute-form in 539 requests, even though HTTP/1.1 clients will only send them in 540 requests to proxies. 542 3.2.3. authority-form 544 The _authority-form_ of request-target is only used for CONNECT 545 requests (Section 9.3.6 of [Semantics]). 547 authority-form = authority 549 When making a CONNECT request to establish a tunnel through one or 550 more proxies, a client MUST send only the target URI's authority 551 component (excluding any userinfo and its "@" delimiter) as the 552 request-target. For example, 554 CONNECT www.example.com:80 HTTP/1.1 556 3.2.4. asterisk-form 558 The _asterisk-form_ of request-target is only used for a server-wide 559 OPTIONS request (Section 9.3.7 of [Semantics]). 561 asterisk-form = "*" 563 When a client wishes to request OPTIONS for the server as a whole, as 564 opposed to a specific named resource of that server, the client MUST 565 send only "*" (%x2A) as the request-target. For example, 566 OPTIONS * HTTP/1.1 568 If a proxy receives an OPTIONS request with an absolute-form of 569 request-target in which the URI has an empty path and no query 570 component, then the last proxy on the request chain MUST send a 571 request-target of "*" when it forwards the request to the indicated 572 origin server. 574 For example, the request 576 OPTIONS http://www.example.org:8001 HTTP/1.1 578 would be forwarded by the final proxy as 580 OPTIONS * HTTP/1.1 581 Host: www.example.org:8001 583 after connecting to port 8001 of host "www.example.org". 585 3.3. Reconstructing the Target URI 587 Since the request-target often contains only part of the user agent's 588 target URI, a server constructs its value to properly service the 589 request (Section 7.1 of [Semantics]). 591 If the request-target is in absolute-form, the target URI is the same 592 as the request-target. Otherwise, the target URI is constructed as 593 follows: 595 o If the server's configuration (or outbound gateway) provides a 596 fixed URI scheme, that scheme is used for the target URI. 597 Otherwise, if the request is received over a secured connection, 598 the target URI's scheme is "https"; if not, the scheme is "http". 600 o If the server's configuration (or outbound gateway) provides a 601 fixed URI authority component, that authority is used for the 602 target URI. If not, then if the request-target is in 603 authority-form, the target URI's authority component is the same 604 as the request-target. If not, then if a Host header field is 605 supplied with a non-empty field value, the authority component is 606 the same as the Host field value. Otherwise, the authority 607 component is assigned the default name configured for the server 608 and, if the connection's incoming TCP port number differs from the 609 default port for the target URI's scheme, then a colon (":") and 610 the incoming port number (in decimal form) are appended to the 611 authority component. 613 o If the request-target is in authority-form or asterisk-form, the 614 target URI's combined path and query component is empty. 615 Otherwise, the combined path and query component is the same as 616 the request-target. 618 o The components of the target URI, once determined as above, can be 619 combined into absolute-URI form by concatenating the scheme, 620 "://", authority, and combined path and query component. 622 Example 1: the following message received over an insecure TCP 623 connection 625 GET /pub/WWW/TheProject.html HTTP/1.1 626 Host: www.example.org:8080 628 has a target URI of 630 http://www.example.org:8080/pub/WWW/TheProject.html 632 Example 2: the following message received over a secured connection 634 OPTIONS * HTTP/1.1 635 Host: www.example.org 637 has a target URI of 639 https://www.example.org 641 Recipients of an HTTP/1.0 request that lacks a Host header field 642 might need to use heuristics (e.g., examination of the URI path for 643 something unique to a particular host) in order to guess the target 644 URI's authority component. 646 4. Status Line 648 The first line of a response message is the status-line, consisting 649 of the protocol version, a space (SP), the status code, another 650 space, and ending with an OPTIONAL textual phrase describing the 651 status code. 653 status-line = HTTP-version SP status-code SP [reason-phrase] 655 Although the status-line grammar rule requires that each of the 656 component elements be separated by a single SP octet, recipients MAY 657 instead parse on whitespace-delimited word boundaries and, aside from 658 the line terminator, treat any form of whitespace as the SP separator 659 while ignoring preceding or trailing whitespace; such whitespace 660 includes one or more of the following octets: SP, HTAB, VT (%x0B), FF 661 (%x0C), or bare CR. However, lenient parsing can result in response 662 splitting security vulnerabilities if there are multiple recipients 663 of the message and each has its own unique interpretation of 664 robustness (see Section 11.1). 666 The status-code element is a 3-digit integer code describing the 667 result of the server's attempt to understand and satisfy the client's 668 corresponding request. The rest of the response message is to be 669 interpreted in light of the semantics defined for that status code. 670 See Section 15 of [Semantics] for information about the semantics of 671 status codes, including the classes of status code (indicated by the 672 first digit), the status codes defined by this specification, 673 considerations for the definition of new status codes, and the IANA 674 registry. 676 status-code = 3DIGIT 678 The reason-phrase element exists for the sole purpose of providing a 679 textual description associated with the numeric status code, mostly 680 out of deference to earlier Internet application protocols that were 681 more frequently used with interactive text clients. 683 reason-phrase = 1*( HTAB / SP / VCHAR / obs-text ) 685 A client SHOULD ignore the reason-phrase content because it is not a 686 reliable channel for information (it might be translated for a given 687 locale, overwritten by intermediaries, or discarded when the message 688 is forwarded via other versions of HTTP). A server MUST send the 689 space that separates status-code from the reason-phrase even when the 690 reason-phrase is absent (i.e., the status-line would end with the 691 three octets SP CR LF). 693 5. Field Syntax 695 Each field line consists of a case-insensitive field name followed by 696 a colon (":"), optional leading whitespace, the field line value, and 697 optional trailing whitespace. 699 field-line = field-name ":" OWS field-value OWS 701 Most HTTP field names and the rules for parsing within field values 702 are defined in Section 6.3 of [Semantics]. This section covers the 703 generic syntax for header field inclusion within, and extraction 704 from, HTTP/1.1 messages. 706 5.1. Field Line Parsing 708 Messages are parsed using a generic algorithm, independent of the 709 individual field names. The contents within a given field line value 710 are not parsed until a later stage of message interpretation (usually 711 after the message's entire field section has been processed). 713 No whitespace is allowed between the field name and colon. In the 714 past, differences in the handling of such whitespace have led to 715 security vulnerabilities in request routing and response handling. A 716 server MUST reject any received request message that contains 717 whitespace between a header field name and colon with a response 718 status code of 400 (Bad Request). A proxy MUST remove any such 719 whitespace from a response message before forwarding the message 720 downstream. 722 A field line value might be preceded and/or followed by optional 723 whitespace (OWS); a single SP preceding the field line value is 724 preferred for consistent readability by humans. The field line value 725 does not include any leading or trailing whitespace: OWS occurring 726 before the first non-whitespace octet of the field line value or 727 after the last non-whitespace octet of the field line value ought to 728 be excluded by parsers when extracting the field line value from a 729 field line. 731 5.2. Obsolete Line Folding 733 Historically, HTTP field line values could be extended over multiple 734 lines by preceding each extra line with at least one space or 735 horizontal tab (obs-fold). This specification deprecates such line 736 folding except within the message/http media type (Section 10.1). 738 obs-fold = OWS CRLF RWS 739 ; obsolete line folding 741 A sender MUST NOT generate a message that includes line folding 742 (i.e., that has any field line value that contains a match to the 743 obs-fold rule) unless the message is intended for packaging within 744 the message/http media type. 746 A server that receives an obs-fold in a request message that is not 747 within a message/http container MUST either reject the message by 748 sending a 400 (Bad Request), preferably with a representation 749 explaining that obsolete line folding is unacceptable, or replace 750 each received obs-fold with one or more SP octets prior to 751 interpreting the field value or forwarding the message downstream. 753 A proxy or gateway that receives an obs-fold in a response message 754 that is not within a message/http container MUST either discard the 755 message and replace it with a 502 (Bad Gateway) response, preferably 756 with a representation explaining that unacceptable line folding was 757 received, or replace each received obs-fold with one or more SP 758 octets prior to interpreting the field value or forwarding the 759 message downstream. 761 A user agent that receives an obs-fold in a response message that is 762 not within a message/http container MUST replace each received 763 obs-fold with one or more SP octets prior to interpreting the field 764 value. 766 6. Message Body 768 The message body (if any) of an HTTP/1.1 message is used to carry 769 payload data (Section 6.4 of [Semantics]) for the request or 770 response. The message body is identical to the payload data unless a 771 transfer coding has been applied, as described in Section 6.1. 773 message-body = *OCTET 775 The rules for determining when a message body is present in an 776 HTTP/1.1 message differ for requests and responses. 778 The presence of a message body in a request is signaled by a 779 Content-Length or Transfer-Encoding header field. Request message 780 framing is independent of method semantics. 782 The presence of a message body in a response depends on both the 783 request method to which it is responding and the response status code 784 (Section 4), and corresponds to when payload data is allowed; see 785 Section 6.4 of [Semantics]. 787 6.1. Transfer-Encoding 789 The Transfer-Encoding header field lists the transfer coding names 790 corresponding to the sequence of transfer codings that have been (or 791 will be) applied to the payload data in order to form the message 792 body. Transfer codings are defined in Section 7. 794 Transfer-Encoding = #transfer-coding 795 ; defined in [Semantics], Section 10.1.4 797 Transfer-Encoding is analogous to the Content-Transfer-Encoding field 798 of MIME, which was designed to enable safe transport of binary data 799 over a 7-bit transport service ([RFC2045], Section 6). However, safe 800 transport has a different focus for an 8bit-clean transfer protocol. 802 In HTTP's case, Transfer-Encoding is primarily intended to accurately 803 delimit a dynamically generated payload and to distinguish payload 804 encodings that are only applied for transport efficiency or security 805 from those that are characteristics of the selected resource. 807 A recipient MUST be able to parse the chunked transfer coding 808 (Section 7.1) because it plays a crucial role in framing messages 809 when the payload data size is not known in advance. A sender MUST 810 NOT apply chunked more than once to a message body (i.e., chunking an 811 already chunked message is not allowed). If any transfer coding 812 other than chunked is applied to a request's payload data, the sender 813 MUST apply chunked as the final transfer coding to ensure that the 814 message is properly framed. If any transfer coding other than 815 chunked is applied to a response's payload data, the sender MUST 816 either apply chunked as the final transfer coding or terminate the 817 message by closing the connection. 819 For example, 821 Transfer-Encoding: gzip, chunked 823 indicates that the payload data has been compressed using the gzip 824 coding and then chunked using the chunked coding while forming the 825 message body. 827 Unlike Content-Encoding (Section 8.5.1 of [Semantics]), Transfer- 828 Encoding is a property of the message, not of the representation, and 829 any recipient along the request/response chain MAY decode the 830 received transfer coding(s) or apply additional transfer coding(s) to 831 the message body, assuming that corresponding changes are made to the 832 Transfer-Encoding field value. Additional information about the 833 encoding parameters can be provided by other header fields not 834 defined by this specification. 836 Transfer-Encoding MAY be sent in a response to a HEAD request or in a 837 304 (Not Modified) response (Section 15.4.5 of [Semantics]) to a GET 838 request, neither of which includes a message body, to indicate that 839 the origin server would have applied a transfer coding to the message 840 body if the request had been an unconditional GET. This indication 841 is not required, however, because any recipient on the response chain 842 (including the origin server) can remove transfer codings when they 843 are not needed. 845 A server MUST NOT send a Transfer-Encoding header field in any 846 response with a status code of 1xx (Informational) or 204 (No 847 Content). A server MUST NOT send a Transfer-Encoding header field in 848 any 2xx (Successful) response to a CONNECT request (Section 9.3.6 of 849 [Semantics]). 851 Transfer-Encoding was added in HTTP/1.1. It is generally assumed 852 that implementations advertising only HTTP/1.0 support will not 853 understand how to process a transfer-encoded payload. A client MUST 854 NOT send a request containing Transfer-Encoding unless it knows the 855 server will handle HTTP/1.1 requests (or later minor revisions); such 856 knowledge might be in the form of specific user configuration or by 857 remembering the version of a prior received response. A server MUST 858 NOT send a response containing Transfer-Encoding unless the 859 corresponding request indicates HTTP/1.1 (or later minor revisions). 861 A server that receives a request message with a transfer coding it 862 does not understand SHOULD respond with 501 (Not Implemented). 864 6.2. Content-Length 866 When a message does not have a Transfer-Encoding header field, a 867 Content-Length header field can provide the anticipated size, as a 868 decimal number of octets, for potential payload data. For messages 869 that do include payload data, the Content-Length field value provides 870 the framing information necessary for determining where the data (and 871 message) ends. For messages that do not include payload data, the 872 Content-Length indicates the size of the selected representation 873 (Section 8.7 of [Semantics]). 875 | *Note:* HTTP's use of Content-Length for message framing 876 | differs significantly from the same field's use in MIME, where 877 | it is an optional field used only within the "message/external- 878 | body" media-type. 880 6.3. Message Body Length 882 The length of a message body is determined by one of the following 883 (in order of precedence): 885 1. Any response to a HEAD request and any response with a 1xx 886 (Informational), 204 (No Content), or 304 (Not Modified) status 887 code is always terminated by the first empty line after the 888 header fields, regardless of the header fields present in the 889 message, and thus cannot contain a message body or trailer 890 section(s). 892 2. Any 2xx (Successful) response to a CONNECT request implies that 893 the connection will become a tunnel immediately after the empty 894 line that concludes the header fields. A client MUST ignore any 895 Content-Length or Transfer-Encoding header fields received in 896 such a message. 898 3. If a Transfer-Encoding header field is present and the chunked 899 transfer coding (Section 7.1) is the final encoding, the message 900 body length is determined by reading and decoding the chunked 901 data until the transfer coding indicates the data is complete. 903 If a Transfer-Encoding header field is present in a response and 904 the chunked transfer coding is not the final encoding, the 905 message body length is determined by reading the connection until 906 it is closed by the server. If a Transfer-Encoding header field 907 is present in a request and the chunked transfer coding is not 908 the final encoding, the message body length cannot be determined 909 reliably; the server MUST respond with the 400 (Bad Request) 910 status code and then close the connection. 912 If a message is received with both a Transfer-Encoding and a 913 Content-Length header field, the Transfer-Encoding overrides the 914 Content-Length. Such a message might indicate an attempt to 915 perform request smuggling (Section 11.2) or response splitting 916 (Section 11.1) and ought to be handled as an error. A sender 917 MUST remove the received Content-Length field prior to forwarding 918 such a message downstream. 920 4. If a message is received without Transfer-Encoding and with an 921 invalid Content-Length header field, then the message framing is 922 invalid and the recipient MUST treat it as an unrecoverable 923 error, unless the field value can be successfully parsed as a 924 comma-separated list (Section 5.6.1 of [Semantics]), all values 925 in the list are valid, and all values in the list are the same. 926 If this is a request message, the server MUST respond with a 400 927 (Bad Request) status code and then close the connection. If this 928 is a response message received by a proxy, the proxy MUST close 929 the connection to the server, discard the received response, and 930 send a 502 (Bad Gateway) response to the client. If this is a 931 response message received by a user agent, the user agent MUST 932 close the connection to the server and discard the received 933 response. 935 5. If a valid Content-Length header field is present without 936 Transfer-Encoding, its decimal value defines the expected message 937 body length in octets. If the sender closes the connection or 938 the recipient times out before the indicated number of octets are 939 received, the recipient MUST consider the message to be 940 incomplete and close the connection. 942 6. If this is a request message and none of the above are true, then 943 the message body length is zero (no message body is present). 945 7. Otherwise, this is a response message without a declared message 946 body length, so the message body length is determined by the 947 number of octets received prior to the server closing the 948 connection. 950 Since there is no way to distinguish a successfully completed, close- 951 delimited response message from a partially received message 952 interrupted by network failure, a server SHOULD generate encoding or 953 length-delimited messages whenever possible. The close-delimiting 954 feature exists primarily for backwards compatibility with HTTP/1.0. 956 | *Note:* Request messages are never close-delimited because they 957 | are always explicitly framed by length or transfer coding, with 958 | the absence of both implying the request ends immediately after 959 | the header section. 961 A server MAY reject a request that contains a message body but not a 962 Content-Length by responding with 411 (Length Required). 964 Unless a transfer coding other than chunked has been applied, a 965 client that sends a request containing a message body SHOULD use a 966 valid Content-Length header field if the message body length is known 967 in advance, rather than the chunked transfer coding, since some 968 existing services respond to chunked with a 411 (Length Required) 969 status code even though they understand the chunked transfer coding. 970 This is typically because such services are implemented via a gateway 971 that requires a content-length in advance of being called and the 972 server is unable or unwilling to buffer the entire request before 973 processing. 975 A user agent that sends a request containing a message body MUST send 976 a valid Content-Length header field if it does not know the server 977 will handle HTTP/1.1 (or later) requests; such knowledge can be in 978 the form of specific user configuration or by remembering the version 979 of a prior received response. 981 If the final response to the last request on a connection has been 982 completely received and there remains additional data to read, a user 983 agent MAY discard the remaining data or attempt to determine if that 984 data belongs as part of the prior message body, which might be the 985 case if the prior message's Content-Length value is incorrect. A 986 client MUST NOT process, cache, or forward such extra data as a 987 separate response, since such behavior would be vulnerable to cache 988 poisoning. 990 7. Transfer Codings 992 Transfer coding names are used to indicate an encoding transformation 993 that has been, can be, or might need to be applied to a payload's 994 data in order to ensure "safe transport" through the network. This 995 differs from a content coding in that the transfer coding is a 996 property of the message rather than a property of the representation 997 that is being transferred. 999 All transfer-coding names are case-insensitive and ought to be 1000 registered within the HTTP Transfer Coding registry, as defined in 1001 Section 7.3. They are used in the Transfer-Encoding (Section 6.1) 1002 and TE (Section 10.1.4 of [Semantics]) header fields (the latter also 1003 defining the "transfer-coding" grammar). 1005 7.1. Chunked Transfer Coding 1007 The chunked transfer coding wraps payload data in order to transfer 1008 it as a series of chunks, each with its own size indicator, followed 1009 by an OPTIONAL trailer section containing trailer fields. Chunked 1010 enables content streams of unknown size to be transferred as a 1011 sequence of length-delimited buffers, which enables the sender to 1012 retain connection persistence and the recipient to know when it has 1013 received the entire message. 1015 chunked-body = *chunk 1016 last-chunk 1017 trailer-section 1018 CRLF 1020 chunk = chunk-size [ chunk-ext ] CRLF 1021 chunk-data CRLF 1022 chunk-size = 1*HEXDIG 1023 last-chunk = 1*("0") [ chunk-ext ] CRLF 1025 chunk-data = 1*OCTET ; a sequence of chunk-size octets 1027 The chunk-size field is a string of hex digits indicating the size of 1028 the chunk-data in octets. The chunked transfer coding is complete 1029 when a chunk with a chunk-size of zero is received, possibly followed 1030 by a trailer section, and finally terminated by an empty line. 1032 A recipient MUST be able to parse and decode the chunked transfer 1033 coding. 1035 Note that HTTP/1.1 does not define any means to limit the size of a 1036 chunked response such that an intermediary can be assured of 1037 buffering the entire response. 1039 The chunked encoding does not define any parameters. Their presence 1040 SHOULD be treated as an error. 1042 7.1.1. Chunk Extensions 1044 The chunked encoding allows each chunk to include zero or more chunk 1045 extensions, immediately following the chunk-size, for the sake of 1046 supplying per-chunk metadata (such as a signature or hash), mid- 1047 message control information, or randomization of message body size. 1049 chunk-ext = *( BWS ";" BWS chunk-ext-name 1050 [ BWS "=" BWS chunk-ext-val ] ) 1052 chunk-ext-name = token 1053 chunk-ext-val = token / quoted-string 1055 The chunked encoding is specific to each connection and is likely to 1056 be removed or recoded by each recipient (including intermediaries) 1057 before any higher-level application would have a chance to inspect 1058 the extensions. Hence, use of chunk extensions is generally limited 1059 to specialized HTTP services such as "long polling" (where client and 1060 server can have shared expectations regarding the use of chunk 1061 extensions) or for padding within an end-to-end secured connection. 1063 A recipient MUST ignore unrecognized chunk extensions. A server 1064 ought to limit the total length of chunk extensions received in a 1065 request to an amount reasonable for the services provided, in the 1066 same way that it applies length limitations and timeouts for other 1067 parts of a message, and generate an appropriate 4xx (Client Error) 1068 response if that amount is exceeded. 1070 7.1.2. Chunked Trailer Section 1072 A trailer section allows the sender to include additional fields at 1073 the end of a chunked message in order to supply metadata that might 1074 be dynamically generated while the payload data is sent, such as a 1075 message integrity check, digital signature, or post-processing 1076 status. The proper use and limitations of trailer fields are defined 1077 in Section 6.5 of [Semantics]. 1079 trailer-section = *( field-line CRLF ) 1081 A recipient that decodes and removes the chunked encoding from a 1082 message (e.g., for storage or forwarding to a non-HTTP/1.1 peer) MUST 1083 discard any received trailer fields, store/forward them separately 1084 from the header fields, or selectively merge into the header section 1085 only those trailer fields corresponding to header field definitions 1086 that are understood by the recipient to explicitly permit and define 1087 how their corresponding trailer field value can be safely merged. 1089 7.1.3. Decoding Chunked 1091 A process for decoding the chunked transfer coding can be represented 1092 in pseudo-code as: 1094 length := 0 1095 read chunk-size, chunk-ext (if any), and CRLF 1096 while (chunk-size > 0) { 1097 read chunk-data and CRLF 1098 append chunk-data to payload-data 1099 length := length + chunk-size 1100 read chunk-size, chunk-ext (if any), and CRLF 1101 } 1102 read trailer field 1103 while (trailer field is not empty) { 1104 if (trailer fields are stored/forwarded separately) { 1105 append trailer field to existing trailer fields 1106 } 1107 else if (trailer field is understood and defined as mergeable) { 1108 merge trailer field with existing header fields 1109 } 1110 else { 1111 discard trailer field 1112 } 1113 read trailer field 1114 } 1115 Content-Length := length 1116 Remove "chunked" from Transfer-Encoding 1117 Remove Trailer from existing header fields 1119 7.2. Transfer Codings for Compression 1121 The following transfer coding names for compression are defined by 1122 the same algorithm as their corresponding content coding: 1124 compress (and x-compress) 1125 See Section 8.5.1.1 of [Semantics]. 1127 deflate 1128 See Section 8.5.1.2 of [Semantics]. 1130 gzip (and x-gzip) 1131 See Section 8.5.1.3 of [Semantics]. 1133 The compression codings do not define any parameters. Their presence 1134 SHOULD be treated as an error. 1136 7.3. Transfer Coding Registry 1138 The "HTTP Transfer Coding Registry" defines the namespace for 1139 transfer coding names. It is maintained at 1140 . 1142 Registrations MUST include the following fields: 1144 o Name 1146 o Description 1148 o Pointer to specification text 1150 Names of transfer codings MUST NOT overlap with names of content 1151 codings (Section 8.5.1 of [Semantics]) unless the encoding 1152 transformation is identical, as is the case for the compression 1153 codings defined in Section 7.2. 1155 The TE header field (Section 10.1.4 of [Semantics]) uses a pseudo 1156 parameter named "q" as rank value when multiple transfer codings are 1157 acceptable. Future registrations of transfer codings SHOULD NOT 1158 define parameters called "q" (case-insensitively) in order to avoid 1159 ambiguities. 1161 Values to be added to this namespace require IETF Review (see 1162 Section 4.8 of [RFC8126]), and MUST conform to the purpose of 1163 transfer coding defined in this specification. 1165 Use of program names for the identification of encoding formats is 1166 not desirable and is discouraged for future encodings. 1168 7.4. Negotiating Transfer Codings 1170 The TE field (Section 10.1.4 of [Semantics]) is used in HTTP/1.1 to 1171 indicate what transfer-codings, besides chunked, the client is 1172 willing to accept in the response, and whether or not the client is 1173 willing to accept trailer fields in a chunked transfer coding. 1175 A client MUST NOT send the chunked transfer coding name in TE; 1176 chunked is always acceptable for HTTP/1.1 recipients. 1178 Three examples of TE use are below. 1180 TE: deflate 1181 TE: 1182 TE: trailers, deflate;q=0.5 1184 When multiple transfer codings are acceptable, the client MAY rank 1185 the codings by preference using a case-insensitive "q" parameter 1186 (similar to the qvalues used in content negotiation fields, 1187 Section 12.4.2 of [Semantics]). The rank value is a real number in 1188 the range 0 through 1, where 0.001 is the least preferred and 1 is 1189 the most preferred; a value of 0 means "not acceptable". 1191 If the TE field value is empty or if no TE field is present, the only 1192 acceptable transfer coding is chunked. A message with no transfer 1193 coding is always acceptable. 1195 The keyword "trailers" indicates that the sender will not discard 1196 trailer fields, as described in Section 6.5 of [Semantics]. 1198 Since the TE header field only applies to the immediate connection, a 1199 sender of TE MUST also send a "TE" connection option within the 1200 Connection header field (Section 7.6.1 of [Semantics]) in order to 1201 prevent the TE header field from being forwarded by intermediaries 1202 that do not support its semantics. 1204 8. Handling Incomplete Messages 1206 A server that receives an incomplete request message, usually due to 1207 a canceled request or a triggered timeout exception, MAY send an 1208 error response prior to closing the connection. 1210 A client that receives an incomplete response message, which can 1211 occur when a connection is closed prematurely or when decoding a 1212 supposedly chunked transfer coding fails, MUST record the message as 1213 incomplete. Cache requirements for incomplete responses are defined 1214 in Section 3 of [Caching]. 1216 If a response terminates in the middle of the header section (before 1217 the empty line is received) and the status code might rely on header 1218 fields to convey the full meaning of the response, then the client 1219 cannot assume that meaning has been conveyed; the client might need 1220 to repeat the request in order to determine what action to take next. 1222 A message body that uses the chunked transfer coding is incomplete if 1223 the zero-sized chunk that terminates the encoding has not been 1224 received. A message that uses a valid Content-Length is incomplete 1225 if the size of the message body received (in octets) is less than the 1226 value given by Content-Length. A response that has neither chunked 1227 transfer coding nor Content-Length is terminated by closure of the 1228 connection and, thus, is considered complete regardless of the number 1229 of message body octets received, provided that the header section was 1230 received intact. 1232 9. Connection Management 1234 HTTP messaging is independent of the underlying transport- or 1235 session-layer connection protocol(s). HTTP only presumes a reliable 1236 transport with in-order delivery of requests and the corresponding 1237 in-order delivery of responses. The mapping of HTTP request and 1238 response structures onto the data units of an underlying transport 1239 protocol is outside the scope of this specification. 1241 As described in Section 7.3 of [Semantics], the specific connection 1242 protocols to be used for an HTTP interaction are determined by client 1243 configuration and the target URI. For example, the "http" URI scheme 1244 (Section 4.2.1 of [Semantics]) indicates a default connection of TCP 1245 over IP, with a default TCP port of 80, but the client might be 1246 configured to use a proxy via some other connection, port, or 1247 protocol. 1249 HTTP implementations are expected to engage in connection management, 1250 which includes maintaining the state of current connections, 1251 establishing a new connection or reusing an existing connection, 1252 processing messages received on a connection, detecting connection 1253 failures, and closing each connection. Most clients maintain 1254 multiple connections in parallel, including more than one connection 1255 per server endpoint. Most servers are designed to maintain thousands 1256 of concurrent connections, while controlling request queues to enable 1257 fair use and detect denial-of-service attacks. 1259 9.1. Establishment 1261 It is beyond the scope of this specification to describe how 1262 connections are established via various transport- or session-layer 1263 protocols. Each connection applies to only one transport link. 1265 9.2. Associating a Response to a Request 1267 HTTP/1.1 does not include a request identifier for associating a 1268 given request message with its corresponding one or more response 1269 messages. Hence, it relies on the order of response arrival to 1270 correspond exactly to the order in which requests are made on the 1271 same connection. More than one response message per request only 1272 occurs when one or more informational responses (1xx, see 1273 Section 15.2 of [Semantics]) precede a final response to the same 1274 request. 1276 A client that has more than one outstanding request on a connection 1277 MUST maintain a list of outstanding requests in the order sent and 1278 MUST associate each received response message on that connection to 1279 the highest ordered request that has not yet received a final (non- 1280 1xx) response. 1282 If an HTTP/1.1 client receives data on a connection that doesn't have 1283 any outstanding requests, it MUST NOT consider them to be a response 1284 to a not-yet-issued request; it SHOULD close the connection, since 1285 message delimitation is now ambiguous, unless the data consists only 1286 of one or more CRLF (which can be discarded, as per Section 2.2). 1288 9.3. Persistence 1290 HTTP/1.1 defaults to the use of "_persistent connections_", allowing 1291 multiple requests and responses to be carried over a single 1292 connection. The "close" connection option is used to signal that a 1293 connection will not persist after the current request/response. HTTP 1294 implementations SHOULD support persistent connections. 1296 A recipient determines whether a connection is persistent or not 1297 based on the most recently received message's protocol version and 1298 Connection header field (Section 7.6.1 of [Semantics]), if any: 1300 o If the "close" connection option is present, the connection will 1301 not persist after the current response; else, 1303 o If the received protocol is HTTP/1.1 (or later), the connection 1304 will persist after the current response; else, 1306 o If the received protocol is HTTP/1.0, the "keep-alive" connection 1307 option is present, either the recipient is not a proxy or the 1308 message is a response, and the recipient wishes to honor the 1309 HTTP/1.0 "keep-alive" mechanism, the connection will persist after 1310 the current response; otherwise, 1312 o The connection will close after the current response. 1314 A client that does not support persistent connections MUST send the 1315 "close" connection option in every request message. 1317 A server that does not support persistent connections MUST send the 1318 "close" connection option in every response message that does not 1319 have a 1xx (Informational) status code. 1321 A client MAY send additional requests on a persistent connection 1322 until it sends or receives a "close" connection option or receives an 1323 HTTP/1.0 response without a "keep-alive" connection option. 1325 In order to remain persistent, all messages on a connection need to 1326 have a self-defined message length (i.e., one not defined by closure 1327 of the connection), as described in Section 6. A server MUST read 1328 the entire request message body or close the connection after sending 1329 its response, since otherwise the remaining data on a persistent 1330 connection would be misinterpreted as the next request. Likewise, a 1331 client MUST read the entire response message body if it intends to 1332 reuse the same connection for a subsequent request. 1334 A proxy server MUST NOT maintain a persistent connection with an 1335 HTTP/1.0 client (see Section 19.7.1 of [RFC2068] for information and 1336 discussion of the problems with the Keep-Alive header field 1337 implemented by many HTTP/1.0 clients). 1339 Note that the field name "Close" is reserved, since using that name 1340 as an HTTP header field might conflict with the "close" connection 1341 defined above. 1343 See Appendix C.2.2 for more information on backwards compatibility 1344 with HTTP/1.0 clients. 1346 9.3.1. Retrying Requests 1348 Connections can be closed at any time, with or without intention. 1349 Implementations ought to anticipate the need to recover from 1350 asynchronous close events. The conditions under which a client can 1351 automatically retry a sequence of outstanding requests are defined in 1352 Section 9.2.2 of [Semantics]. 1354 9.3.2. Pipelining 1356 A client that supports persistent connections MAY "_pipeline_" its 1357 requests (i.e., send multiple requests without waiting for each 1358 response). A server MAY process a sequence of pipelined requests in 1359 parallel if they all have safe methods (Section 9.2.1 of 1360 [Semantics]), but it MUST send the corresponding responses in the 1361 same order that the requests were received. 1363 A client that pipelines requests SHOULD retry unanswered requests if 1364 the connection closes before it receives all of the corresponding 1365 responses. When retrying pipelined requests after a failed 1366 connection (a connection not explicitly closed by the server in its 1367 last complete response), a client MUST NOT pipeline immediately after 1368 connection establishment, since the first remaining request in the 1369 prior pipeline might have caused an error response that can be lost 1370 again if multiple requests are sent on a prematurely closed 1371 connection (see the TCP reset problem described in Section 9.6). 1373 Idempotent methods (Section 9.2.2 of [Semantics]) are significant to 1374 pipelining because they can be automatically retried after a 1375 connection failure. A user agent SHOULD NOT pipeline requests after 1376 a non-idempotent method, until the final response status code for 1377 that method has been received, unless the user agent has a means to 1378 detect and recover from partial failure conditions involving the 1379 pipelined sequence. 1381 An intermediary that receives pipelined requests MAY pipeline those 1382 requests when forwarding them inbound, since it can rely on the 1383 outbound user agent(s) to determine what requests can be safely 1384 pipelined. If the inbound connection fails before receiving a 1385 response, the pipelining intermediary MAY attempt to retry a sequence 1386 of requests that have yet to receive a response if the requests all 1387 have idempotent methods; otherwise, the pipelining intermediary 1388 SHOULD forward any received responses and then close the 1389 corresponding outbound connection(s) so that the outbound user 1390 agent(s) can recover accordingly. 1392 9.4. Concurrency 1394 A client ought to limit the number of simultaneous open connections 1395 that it maintains to a given server. 1397 Previous revisions of HTTP gave a specific number of connections as a 1398 ceiling, but this was found to be impractical for many applications. 1399 As a result, this specification does not mandate a particular maximum 1400 number of connections but, instead, encourages clients to be 1401 conservative when opening multiple connections. 1403 Multiple connections are typically used to avoid the "head-of-line 1404 blocking" problem, wherein a request that takes significant server- 1405 side processing and/or has a large payload blocks subsequent requests 1406 on the same connection. However, each connection consumes server 1407 resources. Furthermore, using multiple connections can cause 1408 undesirable side effects in congested networks. 1410 Note that a server might reject traffic that it deems abusive or 1411 characteristic of a denial-of-service attack, such as an excessive 1412 number of open connections from a single client. 1414 9.5. Failures and Timeouts 1416 Servers will usually have some timeout value beyond which they will 1417 no longer maintain an inactive connection. Proxy servers might make 1418 this a higher value since it is likely that the client will be making 1419 more connections through the same proxy server. The use of 1420 persistent connections places no requirements on the length (or 1421 existence) of this timeout for either the client or the server. 1423 A client or server that wishes to time out SHOULD issue a graceful 1424 close on the connection. Implementations SHOULD constantly monitor 1425 open connections for a received closure signal and respond to it as 1426 appropriate, since prompt closure of both sides of a connection 1427 enables allocated system resources to be reclaimed. 1429 A client, server, or proxy MAY close the transport connection at any 1430 time. For example, a client might have started to send a new request 1431 at the same time that the server has decided to close the "idle" 1432 connection. From the server's point of view, the connection is being 1433 closed while it was idle, but from the client's point of view, a 1434 request is in progress. 1436 A server SHOULD sustain persistent connections, when possible, and 1437 allow the underlying transport's flow-control mechanisms to resolve 1438 temporary overloads, rather than terminate connections with the 1439 expectation that clients will retry. The latter technique can 1440 exacerbate network congestion. 1442 A client sending a message body SHOULD monitor the network connection 1443 for an error response while it is transmitting the request. If the 1444 client sees a response that indicates the server does not wish to 1445 receive the message body and is closing the connection, the client 1446 SHOULD immediately cease transmitting the body and close its side of 1447 the connection. 1449 9.6. Tear-down 1451 The Connection header field (Section 7.6.1 of [Semantics]) provides a 1452 "close" connection option that a sender SHOULD send when it wishes to 1453 close the connection after the current request/response pair. 1455 A client that sends a "close" connection option MUST NOT send further 1456 requests on that connection (after the one containing "close") and 1457 MUST close the connection after reading the final response message 1458 corresponding to this request. 1460 A server that receives a "close" connection option MUST initiate a 1461 close of the connection (see below) after it sends the final response 1462 to the request that contained "close". The server SHOULD send a 1463 "close" connection option in its final response on that connection. 1464 The server MUST NOT process any further requests received on that 1465 connection. 1467 A server that sends a "close" connection option MUST initiate a close 1468 of the connection (see below) after it sends the response containing 1469 "close". The server MUST NOT process any further requests received 1470 on that connection. 1472 A client that receives a "close" connection option MUST cease sending 1473 requests on that connection and close the connection after reading 1474 the response message containing the "close"; if additional pipelined 1475 requests had been sent on the connection, the client SHOULD NOT 1476 assume that they will be processed by the server. 1478 If a server performs an immediate close of a TCP connection, there is 1479 a significant risk that the client will not be able to read the last 1480 HTTP response. If the server receives additional data from the 1481 client on a fully closed connection, such as another request that was 1482 sent by the client before receiving the server's response, the 1483 server's TCP stack will send a reset packet to the client; 1484 unfortunately, the reset packet might erase the client's 1485 unacknowledged input buffers before they can be read and interpreted 1486 by the client's HTTP parser. 1488 To avoid the TCP reset problem, servers typically close a connection 1489 in stages. First, the server performs a half-close by closing only 1490 the write side of the read/write connection. The server then 1491 continues to read from the connection until it receives a 1492 corresponding close by the client, or until the server is reasonably 1493 certain that its own TCP stack has received the client's 1494 acknowledgement of the packet(s) containing the server's last 1495 response. Finally, the server fully closes the connection. 1497 It is unknown whether the reset problem is exclusive to TCP or might 1498 also be found in other transport connection protocols. 1500 Note that a TCP connection that is half-closed by the client does not 1501 delimit a request message, nor does it imply that the client is no 1502 longer interested in a response. In general, transport signals 1503 cannot be relied upon to signal edge cases, since HTTP/1.1 is 1504 independent of transport. 1506 9.7. TLS Connection Initiation 1508 Conceptually, HTTP/TLS is simply sending HTTP messages over a 1509 connection secured via TLS [RFC8446]. 1511 The HTTP client also acts as the TLS client. It initiates a 1512 connection to the server on the appropriate port and sends the TLS 1513 ClientHello to begin the TLS handshake. When the TLS handshake has 1514 finished, the client may then initiate the first HTTP request. All 1515 HTTP data MUST be sent as TLS "application data", but is otherwise 1516 treated like a normal connection for HTTP (including potential reuse 1517 as a persistent connection). 1519 9.8. TLS Connection Closure 1521 TLS provides a facility for secure connection closure. When a valid 1522 closure alert is received, an implementation can be assured that no 1523 further data will be received on that connection. TLS 1524 implementations MUST initiate an exchange of closure alerts before 1525 closing a connection. A TLS implementation MAY, after sending a 1526 closure alert, close the connection without waiting for the peer to 1527 send its closure alert, generating an "incomplete close". Note that 1528 an implementation which does this MAY choose to reuse the session. 1529 This SHOULD only be done when the application knows (typically 1530 through detecting HTTP message boundaries) that it has received all 1531 the message data that it cares about. 1533 As specified in [RFC8446], any implementation which receives a 1534 connection close without first receiving a valid closure alert (a 1535 "premature close") MUST NOT reuse that session. Note that a 1536 premature close does not call into question the security of the data 1537 already received, but simply indicates that subsequent data might 1538 have been truncated. Because TLS is oblivious to HTTP request/ 1539 response boundaries, it is necessary to examine the HTTP data itself 1540 (specifically the Content-Length header) to determine whether the 1541 truncation occurred inside a message or between messages. 1543 When encountering a premature close, a client SHOULD treat as 1544 completed all requests for which it has received as much data as 1545 specified in the Content-Length header. 1547 A client detecting an incomplete close SHOULD recover gracefully. It 1548 MAY resume a TLS session closed in this fashion. 1550 Clients MUST send a closure alert before closing the connection. 1551 Clients which are unprepared to receive any more data MAY choose not 1552 to wait for the server's closure alert and simply close the 1553 connection, thus generating an incomplete close on the server side. 1555 Servers SHOULD be prepared to receive an incomplete close from the 1556 client, since the client can often determine when the end of server 1557 data is. Servers SHOULD be willing to resume TLS sessions closed in 1558 this fashion. 1560 Servers MUST attempt to initiate an exchange of closure alerts with 1561 the client before closing the connection. Servers MAY close the 1562 connection after sending the closure alert, thus generating an 1563 incomplete close on the client side. 1565 10. Enclosing Messages as Data 1567 10.1. Media Type message/http 1569 The message/http media type can be used to enclose a single HTTP 1570 request or response message, provided that it obeys the MIME 1571 restrictions for all "message" types regarding line length and 1572 encodings. 1574 Type name: message 1576 Subtype name: http 1578 Required parameters: N/A 1580 Optional parameters: version, msgtype 1582 version: The HTTP-version number of the enclosed message (e.g., 1583 "1.1"). If not present, the version can be determined from the 1584 first line of the body. 1586 msgtype: The message type - "request" or "response". If not 1587 present, the type can be determined from the first line of the 1588 body. 1590 Encoding considerations: only "7bit", "8bit", or "binary" are 1591 permitted 1593 Security considerations: see Section 11 1594 Interoperability considerations: N/A 1596 Published specification: This specification (see Section 10.1). 1598 Applications that use this media type: N/A 1600 Fragment identifier considerations: N/A 1602 Additional information: Magic number(s): N/A 1604 Deprecated alias names for this type: N/A 1606 File extension(s): N/A 1608 Macintosh file type code(s): N/A 1610 Person and email address to contact for further information: See Aut 1611 hors' Addresses section. 1613 Intended usage: COMMON 1615 Restrictions on usage: N/A 1617 Author: See Authors' Addresses section. 1619 Change controller: IESG 1621 10.2. Media Type application/http 1623 The application/http media type can be used to enclose a pipeline of 1624 one or more HTTP request or response messages (not intermixed). 1626 Type name: application 1628 Subtype name: http 1630 Required parameters: N/A 1632 Optional parameters: version, msgtype 1634 version: The HTTP-version number of the enclosed messages (e.g., 1635 "1.1"). If not present, the version can be determined from the 1636 first line of the body. 1638 msgtype: The message type - "request" or "response". If not 1639 present, the type can be determined from the first line of the 1640 body. 1642 Encoding considerations: HTTP messages enclosed by this type are in 1643 "binary" format; use of an appropriate Content-Transfer-Encoding 1644 is required when transmitted via email. 1646 Security considerations: see Section 11 1648 Interoperability considerations: N/A 1650 Published specification: This specification (see Section 10.2). 1652 Applications that use this media type: N/A 1654 Fragment identifier considerations: N/A 1656 Additional information: Deprecated alias names for this type: N/A 1658 Magic number(s): N/A 1660 File extension(s): N/A 1662 Macintosh file type code(s): N/A 1664 Person and email address to contact for further information: See Aut 1665 hors' Addresses section. 1667 Intended usage: COMMON 1669 Restrictions on usage: N/A 1671 Author: See Authors' Addresses section. 1673 Change controller: IESG 1675 11. Security Considerations 1677 This section is meant to inform developers, information providers, 1678 and users of known security considerations relevant to HTTP message 1679 syntax and parsing. Security considerations about HTTP semantics, 1680 payloads, and routing are addressed in [Semantics]. 1682 11.1. Response Splitting 1684 Response splitting (a.k.a, CRLF injection) is a common technique, 1685 used in various attacks on Web usage, that exploits the line-based 1686 nature of HTTP message framing and the ordered association of 1687 requests to responses on persistent connections [Klein]. This 1688 technique can be particularly damaging when the requests pass through 1689 a shared cache. 1691 Response splitting exploits a vulnerability in servers (usually 1692 within an application server) where an attacker can send encoded data 1693 within some parameter of the request that is later decoded and echoed 1694 within any of the response header fields of the response. If the 1695 decoded data is crafted to look like the response has ended and a 1696 subsequent response has begun, the response has been split and the 1697 content within the apparent second response is controlled by the 1698 attacker. The attacker can then make any other request on the same 1699 persistent connection and trick the recipients (including 1700 intermediaries) into believing that the second half of the split is 1701 an authoritative answer to the second request. 1703 For example, a parameter within the request-target might be read by 1704 an application server and reused within a redirect, resulting in the 1705 same parameter being echoed in the Location header field of the 1706 response. If the parameter is decoded by the application and not 1707 properly encoded when placed in the response field, the attacker can 1708 send encoded CRLF octets and other content that will make the 1709 application's single response look like two or more responses. 1711 A common defense against response splitting is to filter requests for 1712 data that looks like encoded CR and LF (e.g., "%0D" and "%0A"). 1713 However, that assumes the application server is only performing URI 1714 decoding, rather than more obscure data transformations like charset 1715 transcoding, XML entity translation, base64 decoding, sprintf 1716 reformatting, etc. A more effective mitigation is to prevent 1717 anything other than the server's core protocol libraries from sending 1718 a CR or LF within the header section, which means restricting the 1719 output of header fields to APIs that filter for bad octets and not 1720 allowing application servers to write directly to the protocol 1721 stream. 1723 11.2. Request Smuggling 1725 Request smuggling ([Linhart]) is a technique that exploits 1726 differences in protocol parsing among various recipients to hide 1727 additional requests (which might otherwise be blocked or disabled by 1728 policy) within an apparently harmless request. Like response 1729 splitting, request smuggling can lead to a variety of attacks on HTTP 1730 usage. 1732 This specification has introduced new requirements on request 1733 parsing, particularly with regard to message framing in Section 6.3, 1734 to reduce the effectiveness of request smuggling. 1736 11.3. Message Integrity 1738 HTTP does not define a specific mechanism for ensuring message 1739 integrity, instead relying on the error-detection ability of 1740 underlying transport protocols and the use of length or chunk- 1741 delimited framing to detect completeness. Additional integrity 1742 mechanisms, such as hash functions or digital signatures applied to 1743 the content, can be selectively added to messages via extensible 1744 metadata fields. Historically, the lack of a single integrity 1745 mechanism has been justified by the informal nature of most HTTP 1746 communication. However, the prevalence of HTTP as an information 1747 access mechanism has resulted in its increasing use within 1748 environments where verification of message integrity is crucial. 1750 User agents are encouraged to implement configurable means for 1751 detecting and reporting failures of message integrity such that those 1752 means can be enabled within environments for which integrity is 1753 necessary. For example, a browser being used to view medical history 1754 or drug interaction information needs to indicate to the user when 1755 such information is detected by the protocol to be incomplete, 1756 expired, or corrupted during transfer. Such mechanisms might be 1757 selectively enabled via user agent extensions or the presence of 1758 message integrity metadata in a response. At a minimum, user agents 1759 ought to provide some indication that allows a user to distinguish 1760 between a complete and incomplete response message (Section 8) when 1761 such verification is desired. 1763 11.4. Message Confidentiality 1765 HTTP relies on underlying transport protocols to provide message 1766 confidentiality when that is desired. HTTP has been specifically 1767 designed to be independent of the transport protocol, such that it 1768 can be used over many different forms of encrypted connection, with 1769 the selection of such transports being identified by the choice of 1770 URI scheme or within user agent configuration. 1772 The "https" scheme can be used to identify resources that require a 1773 confidential connection, as described in Section 4.2.2 of 1774 [Semantics]. 1776 12. IANA Considerations 1778 The change controller for the following registrations is: "IETF 1779 (iesg@ietf.org) - Internet Engineering Task Force". 1781 12.1. Field Name Registration 1783 First, introduce the new "Hypertext Transfer Protocol (HTTP) Field 1784 Name Registry" at as 1785 described in Section 18.4 of [Semantics]. 1787 Then, please update the registry with the field names listed in the 1788 table below: 1790 ------------------- ---------- ------ ------------ 1791 Field Name Status Ref. Comments 1792 ------------------- ---------- ------ ------------ 1793 Close standard 9.3 (reserved) 1794 MIME-Version standard B.1 1795 Transfer-Encoding standard 6.1 1796 ------------------- ---------- ------ ------------ 1798 Table 1 1800 12.2. Media Type Registration 1802 Please update the "Media Types" registry at 1803 with the registration 1804 information in Section 10.1 and Section 10.2 for the media types 1805 "message/http" and "application/http", respectively. 1807 12.3. Transfer Coding Registration 1809 Please update the "HTTP Transfer Coding Registry" at 1810 with the 1811 registration procedure of Section 7.3 and the content coding names 1812 summarized in the table below. 1814 ------------ ------------------------------- ----------- 1815 Name Description Reference 1816 ------------ ------------------------------- ----------- 1817 chunked Transfer in a series of Section 1818 chunks 7.1 1819 compress UNIX "compress" data format Section 1820 [Welch] 7.2 1821 deflate "deflate" compressed data Section 1822 ([RFC1951]) inside the "zlib" 7.2 1823 data format ([RFC1950]) 1824 gzip GZIP file format [RFC1952] Section 1825 7.2 1826 trailers (reserved) Section 1827 12.3 1828 x-compress Deprecated (alias for Section 1829 compress) 7.2 1830 x-gzip Deprecated (alias for gzip) Section 1831 7.2 1832 ------------ ------------------------------- ----------- 1834 Table 2 1836 | *Note:* the coding name "trailers" is reserved because its use 1837 | would conflict with the keyword "trailers" in the TE header 1838 | field (Section 10.1.4 of [Semantics]). 1840 12.4. ALPN Protocol ID Registration 1842 Please update the "TLS Application-Layer Protocol Negotiation (ALPN) 1843 Protocol IDs" registry at with the 1845 registration below: 1847 ---------- ----------------------------- ---------------- 1848 Protocol Identification Sequence Reference 1849 ---------- ----------------------------- ---------------- 1850 HTTP/1.1 0x68 0x74 0x74 0x70 0x2f (this 1851 0x31 0x2e 0x31 ("http/1.1") specification) 1852 ---------- ----------------------------- ---------------- 1854 Table 3 1856 13. References 1858 13.1. Normative References 1860 [Caching] Fielding, R., Ed., Nottingham, M., Ed., and J. Reschke, 1861 Ed., "HTTP Caching", Work in Progress, Internet-Draft, 1862 draft-ietf-httpbis-cache-13, December 14, 2020, 1863 . 1865 [RFC1950] Deutsch, L.P. and J-L. Gailly, "ZLIB Compressed Data 1866 Format Specification version 3.3", RFC 1950, 1867 DOI 10.17487/RFC1950, May 1996, 1868 . 1870 [RFC1951] Deutsch, P., "DEFLATE Compressed Data Format Specification 1871 version 1.3", RFC 1951, DOI 10.17487/RFC1951, May 1996, 1872 . 1874 [RFC1952] Deutsch, P., Gailly, J-L., Adler, M., Deutsch, L.P., and 1875 G. Randers-Pehrson, "GZIP file format specification 1876 version 4.3", RFC 1952, DOI 10.17487/RFC1952, May 1996, 1877 . 1879 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 1880 Requirement Levels", BCP 14, RFC 2119, 1881 DOI 10.17487/RFC2119, March 1997, 1882 . 1884 [RFC3986] Berners-Lee, T., Fielding, R., and L. Masinter, "Uniform 1885 Resource Identifier (URI): Generic Syntax", STD 66, 1886 RFC 3986, DOI 10.17487/RFC3986, January 2005, 1887 . 1889 [RFC5234] Crocker, D., Ed. and P. Overell, "Augmented BNF for Syntax 1890 Specifications: ABNF", STD 68, RFC 5234, 1891 DOI 10.17487/RFC5234, January 2008, 1892 . 1894 [RFC7405] Kyzivat, P., "Case-Sensitive String Support in ABNF", 1895 RFC 7405, DOI 10.17487/RFC7405, December 2014, 1896 . 1898 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 1899 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 1900 May 2017, . 1902 [RFC8446] Rescorla, E., "The Transport Layer Security (TLS) Protocol 1903 Version 1.3", RFC 8446, DOI 10.17487/RFC8446, August 2018, 1904 . 1906 [Semantics] 1907 Fielding, R., Ed., Nottingham, M., Ed., and J. Reschke, 1908 Ed., "HTTP Semantics", Work in Progress, Internet-Draft, 1909 draft-ietf-httpbis-semantics-13, December 14, 2020, 1910 . 1913 [USASCII] American National Standards Institute, "Coded Character 1914 Set -- 7-bit American Standard Code for Information 1915 Interchange", ANSI X3.4, 1986. 1917 [Welch] Welch, T. A., "A Technique for High-Performance Data 1918 Compression", IEEE Computer 17(6), June 1984. 1920 13.2. Informative References 1922 [Err4667] RFC Errata, Erratum ID 4667, RFC 7230, 1923 . 1925 [Klein] Klein, A., "Divide and Conquer - HTTP Response Splitting, 1926 Web Cache Poisoning Attacks, and Related Topics", March 1927 2004, . 1930 [Linhart] Linhart, C., Klein, A., Heled, R., and S. Orrin, "HTTP 1931 Request Smuggling", June 2005, 1932 . 1935 [RFC1945] Berners-Lee, T., Fielding, R.T., and H.F. Nielsen, 1936 "Hypertext Transfer Protocol -- HTTP/1.0", RFC 1945, 1937 DOI 10.17487/RFC1945, May 1996, 1938 . 1940 [RFC2045] Freed, N. and N.S. Borenstein, "Multipurpose Internet Mail 1941 Extensions (MIME) Part One: Format of Internet Message 1942 Bodies", RFC 2045, DOI 10.17487/RFC2045, November 1996, 1943 . 1945 [RFC2046] Freed, N. and N. Borenstein, "Multipurpose Internet Mail 1946 Extensions (MIME) Part Two: Media Types", RFC 2046, 1947 DOI 10.17487/RFC2046, November 1996, 1948 . 1950 [RFC2049] Freed, N. and N.S. Borenstein, "Multipurpose Internet Mail 1951 Extensions (MIME) Part Five: Conformance Criteria and 1952 Examples", RFC 2049, DOI 10.17487/RFC2049, November 1996, 1953 . 1955 [RFC2068] Fielding, R., Gettys, J., Mogul, J., Nielsen, H., and T. 1956 Berners-Lee, "Hypertext Transfer Protocol -- HTTP/1.1", 1957 RFC 2068, DOI 10.17487/RFC2068, January 1997, 1958 . 1960 [RFC2557] Palme, F., Hopmann, A., Shelness, N., and E. Stefferud, 1961 "MIME Encapsulation of Aggregate Documents, such as HTML 1962 (MHTML)", RFC 2557, DOI 10.17487/RFC2557, March 1999, 1963 . 1965 [RFC5322] Resnick, P., "Internet Message Format", RFC 5322, 1966 DOI 10.17487/RFC5322, October 2008, 1967 . 1969 [RFC7230] Fielding, R., Ed. and J. F. Reschke, Ed., "Hypertext 1970 Transfer Protocol (HTTP/1.1): Message Syntax and Routing", 1971 RFC 7230, DOI 10.17487/RFC7230, June 2014, 1972 . 1974 [RFC7231] Fielding, R., Ed. and J. F. Reschke, Ed., "Hypertext 1975 Transfer Protocol (HTTP/1.1): Semantics and Content", 1976 RFC 7231, DOI 10.17487/RFC7231, June 2014, 1977 . 1979 [RFC8126] Cotton, M., Leiba, B., and T. Narten, "Guidelines for 1980 Writing an IANA Considerations Section in RFCs", BCP 26, 1981 RFC 8126, DOI 10.17487/RFC8126, June 2017, 1982 . 1984 Appendix A. Collected ABNF 1986 In the collected ABNF below, list rules are expanded as per 1987 Section 5.6.1.1 of [Semantics]. 1989 BWS = 1991 HTTP-message = start-line CRLF *( field-line CRLF ) CRLF [ 1992 message-body ] 1993 HTTP-name = %x48.54.54.50 ; HTTP 1994 HTTP-version = HTTP-name "/" DIGIT "." DIGIT 1996 OWS = 1998 RWS = 2000 Transfer-Encoding = [ transfer-coding *( OWS "," OWS transfer-coding 2001 ) ] 2003 absolute-URI = 2004 absolute-form = absolute-URI 2005 absolute-path = 2006 asterisk-form = "*" 2007 authority = 2008 authority-form = authority 2010 chunk = chunk-size [ chunk-ext ] CRLF chunk-data CRLF 2011 chunk-data = 1*OCTET 2012 chunk-ext = *( BWS ";" BWS chunk-ext-name [ BWS "=" BWS chunk-ext-val 2013 ] ) 2014 chunk-ext-name = token 2015 chunk-ext-val = token / quoted-string 2016 chunk-size = 1*HEXDIG 2017 chunked-body = *chunk last-chunk trailer-section CRLF 2018 comment = 2020 field-line = field-name ":" OWS field-value OWS 2021 field-name = 2022 field-value = 2024 last-chunk = 1*"0" [ chunk-ext ] CRLF 2026 message-body = *OCTET 2027 method = token 2029 obs-fold = OWS CRLF RWS 2030 obs-text = 2031 origin-form = absolute-path [ "?" query ] 2033 port = 2035 query = 2036 quoted-string = 2038 reason-phrase = 1*( HTAB / SP / VCHAR / obs-text ) 2039 request-line = method SP request-target SP HTTP-version 2040 request-target = origin-form / absolute-form / authority-form / 2041 asterisk-form 2043 start-line = request-line / status-line 2044 status-code = 3DIGIT 2045 status-line = HTTP-version SP status-code SP [ reason-phrase ] 2047 token = 2048 trailer-section = *( field-line CRLF ) 2049 transfer-coding = 2050 uri-host = 2052 Appendix B. Differences between HTTP and MIME 2054 HTTP/1.1 uses many of the constructs defined for the Internet Message 2055 Format [RFC5322] and the Multipurpose Internet Mail Extensions (MIME) 2056 [RFC2045] to allow a message body to be transmitted in an open 2057 variety of representations and with extensible fields. However, RFC 2058 2045 is focused only on email; applications of HTTP have many 2059 characteristics that differ from email; hence, HTTP has features that 2060 differ from MIME. These differences were carefully chosen to 2061 optimize performance over binary connections, to allow greater 2062 freedom in the use of new media types, to make date comparisons 2063 easier, and to acknowledge the practice of some early HTTP servers 2064 and clients. 2066 This appendix describes specific areas where HTTP differs from MIME. 2067 Proxies and gateways to and from strict MIME environments need to be 2068 aware of these differences and provide the appropriate conversions 2069 where necessary. 2071 B.1. MIME-Version 2073 HTTP is not a MIME-compliant protocol. However, messages can include 2074 a single MIME-Version header field to indicate what version of the 2075 MIME protocol was used to construct the message. Use of the MIME- 2076 Version header field indicates that the message is in full 2077 conformance with the MIME protocol (as defined in [RFC2045]). 2078 Senders are responsible for ensuring full conformance (where 2079 possible) when exporting HTTP messages to strict MIME environments. 2081 B.2. Conversion to Canonical Form 2083 MIME requires that an Internet mail body part be converted to 2084 canonical form prior to being transferred, as described in Section 4 2085 of [RFC2049]. Section 8.4.3 of [Semantics] describes the forms 2086 allowed for subtypes of the "text" media type when transmitted over 2087 HTTP. [RFC2046] requires that content with a type of "text" 2088 represent line breaks as CRLF and forbids the use of CR or LF outside 2089 of line break sequences. HTTP allows CRLF, bare CR, and bare LF to 2090 indicate a line break within text content. 2092 A proxy or gateway from HTTP to a strict MIME environment ought to 2093 translate all line breaks within text media types to the RFC 2049 2094 canonical form of CRLF. Note, however, this might be complicated by 2095 the presence of a Content-Encoding and by the fact that HTTP allows 2096 the use of some charsets that do not use octets 13 and 10 to 2097 represent CR and LF, respectively. 2099 Conversion will break any cryptographic checksums applied to the 2100 original content unless the original content is already in canonical 2101 form. Therefore, the canonical form is recommended for any content 2102 that uses such checksums in HTTP. 2104 B.3. Conversion of Date Formats 2106 HTTP/1.1 uses a restricted set of date formats (Section 5.6.7 of 2107 [Semantics]) to simplify the process of date comparison. Proxies and 2108 gateways from other protocols ought to ensure that any Date header 2109 field present in a message conforms to one of the HTTP/1.1 formats 2110 and rewrite the date if necessary. 2112 B.4. Conversion of Content-Encoding 2114 MIME does not include any concept equivalent to HTTP/1.1's Content- 2115 Encoding header field. Since this acts as a modifier on the media 2116 type, proxies and gateways from HTTP to MIME-compliant protocols 2117 ought to either change the value of the Content-Type header field or 2118 decode the representation before forwarding the message. (Some 2119 experimental applications of Content-Type for Internet mail have used 2120 a media-type parameter of ";conversions=" to perform 2121 a function equivalent to Content-Encoding. However, this parameter 2122 is not part of the MIME standards). 2124 B.5. Conversion of Content-Transfer-Encoding 2126 HTTP does not use the Content-Transfer-Encoding field of MIME. 2127 Proxies and gateways from MIME-compliant protocols to HTTP need to 2128 remove any Content-Transfer-Encoding prior to delivering the response 2129 message to an HTTP client. 2131 Proxies and gateways from HTTP to MIME-compliant protocols are 2132 responsible for ensuring that the message is in the correct format 2133 and encoding for safe transport on that protocol, where "safe 2134 transport" is defined by the limitations of the protocol being used. 2135 Such a proxy or gateway ought to transform and label the data with an 2136 appropriate Content-Transfer-Encoding if doing so will improve the 2137 likelihood of safe transport over the destination protocol. 2139 B.6. MHTML and Line Length Limitations 2141 HTTP implementations that share code with MHTML [RFC2557] 2142 implementations need to be aware of MIME line length limitations. 2143 Since HTTP does not have this limitation, HTTP does not fold long 2144 lines. MHTML messages being transported by HTTP follow all 2145 conventions of MHTML, including line length limitations and folding, 2146 canonicalization, etc., since HTTP transfers message-bodies as 2147 payload and, aside from the "multipart/byteranges" type (Section 14.5 2148 of [Semantics]), does not interpret the content or any MIME header 2149 lines that might be contained therein. 2151 Appendix C. Changes from previous RFCs 2153 C.1. Changes from HTTP/0.9 2155 Since HTTP/0.9 did not support header fields in a request, there is 2156 no mechanism for it to support name-based virtual hosts (selection of 2157 resource by inspection of the Host header field). Any server that 2158 implements name-based virtual hosts ought to disable support for 2159 HTTP/0.9. Most requests that appear to be HTTP/0.9 are, in fact, 2160 badly constructed HTTP/1.x requests caused by a client failing to 2161 properly encode the request-target. 2163 C.2. Changes from HTTP/1.0 2165 C.2.1. Multihomed Web Servers 2167 The requirements that clients and servers support the Host header 2168 field (Section 7.2 of [Semantics]), report an error if it is missing 2169 from an HTTP/1.1 request, and accept absolute URIs (Section 3.2) are 2170 among the most important changes defined by HTTP/1.1. 2172 Older HTTP/1.0 clients assumed a one-to-one relationship of IP 2173 addresses and servers; there was no other established mechanism for 2174 distinguishing the intended server of a request than the IP address 2175 to which that request was directed. The Host header field was 2176 introduced during the development of HTTP/1.1 and, though it was 2177 quickly implemented by most HTTP/1.0 browsers, additional 2178 requirements were placed on all HTTP/1.1 requests in order to ensure 2179 complete adoption. At the time of this writing, most HTTP-based 2180 services are dependent upon the Host header field for targeting 2181 requests. 2183 C.2.2. Keep-Alive Connections 2185 In HTTP/1.0, each connection is established by the client prior to 2186 the request and closed by the server after sending the response. 2187 However, some implementations implement the explicitly negotiated 2188 ("Keep-Alive") version of persistent connections described in 2189 Section 19.7.1 of [RFC2068]. 2191 Some clients and servers might wish to be compatible with these 2192 previous approaches to persistent connections, by explicitly 2193 negotiating for them with a "Connection: keep-alive" request header 2194 field. However, some experimental implementations of HTTP/1.0 2195 persistent connections are faulty; for example, if an HTTP/1.0 proxy 2196 server doesn't understand Connection, it will erroneously forward 2197 that header field to the next inbound server, which would result in a 2198 hung connection. 2200 One attempted solution was the introduction of a Proxy-Connection 2201 header field, targeted specifically at proxies. In practice, this 2202 was also unworkable, because proxies are often deployed in multiple 2203 layers, bringing about the same problem discussed above. 2205 As a result, clients are encouraged not to send the Proxy-Connection 2206 header field in any requests. 2208 Clients are also encouraged to consider the use of Connection: keep- 2209 alive in requests carefully; while they can enable persistent 2210 connections with HTTP/1.0 servers, clients using them will need to 2211 monitor the connection for "hung" requests (which indicate that the 2212 client ought stop sending the header field), and this mechanism ought 2213 not be used by clients at all when a proxy is being used. 2215 C.2.3. Introduction of Transfer-Encoding 2217 HTTP/1.1 introduces the Transfer-Encoding header field (Section 6.1). 2218 Transfer codings need to be decoded prior to forwarding an HTTP 2219 message over a MIME-compliant protocol. 2221 C.3. Changes from RFC 7230 2223 Most of the sections introducing HTTP's design goals, history, 2224 architecture, conformance criteria, protocol versioning, URIs, 2225 message routing, and header fields have been moved to [Semantics]. 2226 This document has been reduced to just the messaging syntax and 2227 connection management requirements specific to HTTP/1.1. 2229 Prohibited generation of bare CRs outside of payload data. 2230 (Section 2.2) 2232 In the ABNF for chunked extensions, re-introduced (bad) whitespace 2233 around ";" and "=". Whitespace was removed in [RFC7230], but that 2234 change was found to break existing implementations (see [Err4667]). 2235 (Section 7.1.1) 2237 Trailer field semantics now transcend the specifics of chunked 2238 encoding. The decoding algorithm for chunked (Section 7.1.3) has 2239 been updated to encourage storage/forwarding of trailer fields 2240 separately from the header section, to only allow merging into the 2241 header section if the recipient knows the corresponding field 2242 definition permits and defines how to merge, and otherwise to discard 2243 the trailer fields instead of merging. The trailer part is now 2244 called the trailer section to be more consistent with the header 2245 section and more distinct from a body part. (Section 7.1.2) 2247 Disallowed transfer coding parameters called "q" in order to avoid 2248 conflicts with the use of ranks in the TE header field. 2249 (Section 7.3) 2251 Appendix D. Change Log 2253 This section is to be removed before publishing as an RFC. 2255 D.1. Between RFC7230 and draft 00 2257 The changes were purely editorial: 2259 o Change boilerplate and abstract to indicate the "draft" status, 2260 and update references to ancestor specifications. 2262 o Adjust historical notes. 2264 o Update links to sibling specifications. 2266 o Replace sections listing changes from RFC 2616 by new empty 2267 sections referring to RFC 723x. 2269 o Remove acknowledgements specific to RFC 723x. 2271 o Move "Acknowledgements" to the very end and make them unnumbered. 2273 D.2. Since draft-ietf-httpbis-messaging-00 2275 The changes in this draft are editorial, with respect to HTTP as a 2276 whole, to move all core HTTP semantics into [Semantics]: 2278 o Moved introduction, architecture, conformance, and ABNF extensions 2279 from RFC 7230 (Messaging) to semantics [Semantics]. 2281 o Moved discussion of MIME differences from RFC 7231 (Semantics) to 2282 Appendix B since they mostly cover transforming 1.1 messages. 2284 o Moved all extensibility tips, registration procedures, and 2285 registry tables from the IANA considerations to normative 2286 sections, reducing the IANA considerations to just instructions 2287 that will be removed prior to publication as an RFC. 2289 D.3. Since draft-ietf-httpbis-messaging-01 2290 o Cite RFC 8126 instead of RFC 5226 () 2293 o Resolved erratum 4779, no change needed here 2294 (, 2295 ) 2297 o In Section 7, fixed prose claiming transfer parameters allow bare 2298 names (, 2299 ) 2301 o Resolved erratum 4225, no change needed here 2302 (, 2303 ) 2305 o Replace "response code" with "response status code" 2306 (, 2307 ) 2309 o In Section 9.3, clarify statement about HTTP/1.0 keep-alive 2310 (, 2311 ) 2313 o In Section 7.1.1, re-introduce (bad) whitespace around ";" and "=" 2314 (, 2315 , ) 2318 o In Section 7.3, state that transfer codings should not use 2319 parameters named "q" (, ) 2322 o In Section 7, mark coding name "trailers" as reserved in the IANA 2323 registry () 2325 D.4. Since draft-ietf-httpbis-messaging-02 2327 o In Section 4, explain why the reason phrase should be ignored by 2328 clients (). 2330 o Add Section 9.2 to explain how request/response correlation is 2331 performed () 2333 D.5. Since draft-ietf-httpbis-messaging-03 2335 o In Section 9.2, caution against treating data on a connection as 2336 part of a not-yet-issued request () 2339 o In Section 7, remove the predefined codings from the ABNF and make 2340 it generic instead () 2343 o Use RFC 7405 ABNF notation for case-sensitive string constants 2344 () 2346 D.6. Since draft-ietf-httpbis-messaging-04 2348 o In Section 7.8 of [Semantics], clarify that protocol-name is to be 2349 matched case-insensitively () 2352 o In Section 5.2, add leading optional whitespace to obs-fold ABNF 2353 (, 2354 ) 2356 o In Section 4, add clarifications about empty reason phrases 2357 () 2359 o Move discussion of retries from Section 9.3.1 into [Semantics] 2360 () 2362 D.7. Since draft-ietf-httpbis-messaging-05 2364 o In Section 7.1.2, the trailer part has been renamed the trailer 2365 section (for consistency with the header section) and trailers are 2366 no longer merged as header fields by default, but rather can be 2367 discarded, kept separate from header fields, or merged with header 2368 fields only if understood and defined as being mergeable 2369 () 2371 o In Section 2.1 and related Sections, move the trailing CRLF from 2372 the line grammars into the message format 2373 () 2375 o Moved Section 2.3 down () 2378 o In Section 7.8 of [Semantics], use 'websocket' instead of 2379 'HTTP/2.0' in examples () 2382 o Move version non-specific text from Section 6 into semantics as 2383 "payload" () 2385 o In Section 9.8, add text from RFC 2818 2386 () 2388 D.8. Since draft-ietf-httpbis-messaging-06 2390 o In Section 12.4, update the APLN protocol id for HTTP/1.1 2391 () 2393 o In Section 5, align with updates to field terminology in semantics 2394 () 2396 o In Section 7.6.1 of [Semantics], clarify that new connection 2397 options indeed need to be registered () 2400 o In Section 1.1, reference RFC 8174 as well 2401 () 2403 D.9. Since draft-ietf-httpbis-messaging-07 2405 o Move TE: trailers into [Semantics] () 2408 o In Section 6.3, adjust requirements for handling multiple content- 2409 length values () 2411 o Throughout, replace "effective request URI" with "target URI" 2412 () 2414 o In Section 6.1, don't claim Transfer-Encoding is supported by 2415 HTTP/2 or later () 2417 D.10. Since draft-ietf-httpbis-messaging-08 2419 o In Section 2.2, disallow bare CRs () 2422 o Appendix A now uses the sender variant of the "#" list expansion 2423 () 2425 o In Section 5, adjust IANA "Close" entry for new registry format 2426 () 2428 D.11. Since draft-ietf-httpbis-messaging-09 2430 o Switch to xml2rfc v3 mode for draft generation 2431 () 2433 D.12. Since draft-ietf-httpbis-messaging-10 2434 o In Section 6.3, note that TCP half-close does not delimit a 2435 request; talk about corresponding server-side behaviour in 2436 Section 9.6 () 2438 o Moved requirements specific to HTTP/1.1 from [Semantics] into 2439 Section 3.2 () 2441 o In Section 6.1 (Transfer-Encoding), adjust ABNF to allow empty 2442 lists () 2444 o In Section 9.7, add text from RFC 2818 2445 () 2447 o Moved definitions of "TE" and "Upgrade" into [Semantics] 2448 () 2450 o Moved definition of "Connection" into [Semantics] 2451 () 2453 D.13. Since draft-ietf-httpbis-messaging-11 2455 o Move IANA Upgrade Token Registry instructions to [Semantics] 2456 () 2458 D.14. Since draft-ietf-httpbis-messaging-12 2460 o Moved content of history appendix to Semantics 2461 () 2463 o Moved note about "close" being reserved as field name to 2464 Section 9.3 () 2466 o Moved table of transfer codings into Section 12.3 2467 () 2469 o In Section 13.2, updated the URI for the [Linhart] paper 2470 () 2472 o Changed document title to just "HTTP/1.1" 2473 () 2475 o In Section 7, moved transfer-coding ABNF to Section 10.1.4 of 2476 [Semantics] () 2478 o Changed to using "payload data" when defining requirements about 2479 the data being conveyed within a message, instead of the terms 2480 "payload body" or "response body" or "representation body", since 2481 they often get confused with the HTTP/1.1 message body (which 2482 includes transfer coding) () 2485 Acknowledgments 2487 See Appendix "Acknowledgments" of [Semantics]. 2489 Authors' Addresses 2491 Roy T. Fielding (editor) 2492 Adobe 2493 345 Park Ave 2494 San Jose, CA 95110 2495 United States of America 2497 Email: fielding@gbiv.com 2498 URI: https://roy.gbiv.com/ 2500 Mark Nottingham (editor) 2501 Fastly 2502 Prahran VIC 2503 Australia 2505 Email: mnot@mnot.net 2506 URI: https://www.mnot.net/ 2508 Julian Reschke (editor) 2509 greenbytes GmbH 2510 Hafenweg 16 2511 48155 Münster 2512 Germany 2514 Email: julian.reschke@greenbytes.de 2515 URI: https://greenbytes.de/tech/webdav/