idnits 2.17.1 draft-ietf-httpbis-messaging-16.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 == 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 (27 May 2021) is 1058 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in RFCs) == Unused Reference: 'RFC7231' is defined on line 2034, but no explicit reference was found in the text == Outdated reference: A later version (-19) exists of draft-ietf-httpbis-cache-16 -- Possible downref: Normative reference to a draft: ref.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reference to an Informational RFC: RFC 1951 ** Downref: Normative reference to an Informational RFC: RFC 1952 == Outdated reference: A later version (-19) exists of draft-ietf-httpbis-semantics-16 -- Possible downref: Normative reference to a draft: ref.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 7230 (ref. 'Err4667') (Obsoleted by RFC 9110, RFC 9112) -- Obsolete informational reference (is this intentional?): RFC 2068 (Obsoleted by RFC 2616) -- Duplicate reference: RFC7230, mentioned in 'RFC7230', was also mentioned in 'Err4667'. -- Obsolete informational reference (is this intentional?): RFC 7230 (Obsoleted by RFC 9110, RFC 9112) -- Obsolete informational reference (is this intentional?): RFC 7231 (Obsoleted by RFC 9110) Summary: 3 errors (**), 0 flaws (~~), 6 warnings (==), 11 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: 28 November 2021 J. Reschke, Ed. 7 greenbytes 8 27 May 2021 10 HTTP/1.1 11 draft-ietf-httpbis-messaging-16 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.17. 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 28 November 2021. 54 Copyright Notice 56 Copyright (c) 2021 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 . . . . . . . . . . . . . . . . . . . . 13 96 3.3. Reconstructing the Target URI . . . . . . . . . . . . . . 13 97 4. Status Line . . . . . . . . . . . . . . . . . . . . . . . . . 15 98 5. Field Syntax . . . . . . . . . . . . . . . . . . . . . . . . 16 99 5.1. Field Line Parsing . . . . . . . . . . . . . . . . . . . 16 100 5.2. Obsolete Line Folding . . . . . . . . . . . . . . . . . . 17 101 6. Message Body . . . . . . . . . . . . . . . . . . . . . . . . 17 102 6.1. Transfer-Encoding . . . . . . . . . . . . . . . . . . . . 18 103 6.2. Content-Length . . . . . . . . . . . . . . . . . . . . . 19 104 6.3. Message Body Length . . . . . . . . . . . . . . . . . . . 20 105 7. Transfer Codings . . . . . . . . . . . . . . . . . . . . . . 22 106 7.1. Chunked Transfer Coding . . . . . . . . . . . . . . . . . 23 107 7.1.1. Chunk Extensions . . . . . . . . . . . . . . . . . . 23 108 7.1.2. Chunked Trailer Section . . . . . . . . . . . . . . . 24 109 7.1.3. Decoding Chunked . . . . . . . . . . . . . . . . . . 24 110 7.2. Transfer Codings for Compression . . . . . . . . . . . . 25 111 7.3. Transfer Coding Registry . . . . . . . . . . . . . . . . 25 112 7.4. Negotiating Transfer Codings . . . . . . . . . . . . . . 26 113 8. Handling Incomplete Messages . . . . . . . . . . . . . . . . 27 114 9. Connection Management . . . . . . . . . . . . . . . . . . . . 28 115 9.1. Establishment . . . . . . . . . . . . . . . . . . . . . . 28 116 9.2. Associating a Response to a Request . . . . . . . . . . . 28 117 9.3. Persistence . . . . . . . . . . . . . . . . . . . . . . . 29 118 9.3.1. Retrying Requests . . . . . . . . . . . . . . . . . . 30 119 9.3.2. Pipelining . . . . . . . . . . . . . . . . . . . . . 30 120 9.4. Concurrency . . . . . . . . . . . . . . . . . . . . . . . 31 121 9.5. Failures and Timeouts . . . . . . . . . . . . . . . . . . 31 122 9.6. Tear-down . . . . . . . . . . . . . . . . . . . . . . . . 32 123 9.7. TLS Connection Initiation . . . . . . . . . . . . . . . . 34 124 9.8. TLS Connection Closure . . . . . . . . . . . . . . . . . 34 125 10. Enclosing Messages as Data . . . . . . . . . . . . . . . . . 35 126 10.1. Media Type message/http . . . . . . . . . . . . . . . . 35 127 10.2. Media Type application/http . . . . . . . . . . . . . . 36 128 11. Security Considerations . . . . . . . . . . . . . . . . . . . 37 129 11.1. Response Splitting . . . . . . . . . . . . . . . . . . . 37 130 11.2. Request Smuggling . . . . . . . . . . . . . . . . . . . 38 131 11.3. Message Integrity . . . . . . . . . . . . . . . . . . . 39 132 11.4. Message Confidentiality . . . . . . . . . . . . . . . . 39 133 12. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 40 134 12.1. Field Name Registration . . . . . . . . . . . . . . . . 40 135 12.2. Media Type Registration . . . . . . . . . . . . . . . . 40 136 12.3. Transfer Coding Registration . . . . . . . . . . . . . . 40 137 12.4. ALPN Protocol ID Registration . . . . . . . . . . . . . 41 138 13. References . . . . . . . . . . . . . . . . . . . . . . . . . 42 139 13.1. Normative References . . . . . . . . . . . . . . . . . . 42 140 13.2. Informative References . . . . . . . . . . . . . . . . . 43 141 Appendix A. Collected ABNF . . . . . . . . . . . . . . . . . . . 44 142 Appendix B. Differences between HTTP and MIME . . . . . . . . . 46 143 B.1. MIME-Version . . . . . . . . . . . . . . . . . . . . . . 46 144 B.2. Conversion to Canonical Form . . . . . . . . . . . . . . 46 145 B.3. Conversion of Date Formats . . . . . . . . . . . . . . . 47 146 B.4. Conversion of Content-Encoding . . . . . . . . . . . . . 47 147 B.5. Conversion of Content-Transfer-Encoding . . . . . . . . . 47 148 B.6. MHTML and Line Length Limitations . . . . . . . . . . . . 48 149 Appendix C. Changes from previous RFCs . . . . . . . . . . . . . 48 150 C.1. Changes from HTTP/0.9 . . . . . . . . . . . . . . . . . . 48 151 C.2. Changes from HTTP/1.0 . . . . . . . . . . . . . . . . . . 48 152 C.2.1. Multihomed Web Servers . . . . . . . . . . . . . . . 48 153 C.2.2. Keep-Alive Connections . . . . . . . . . . . . . . . 49 154 C.2.3. Introduction of Transfer-Encoding . . . . . . . . . . 49 155 C.3. Changes from RFC 7230 . . . . . . . . . . . . . . . . . . 49 156 Appendix D. Change Log . . . . . . . . . . . . . . . . . . . . . 50 157 D.1. Between RFC7230 and draft 00 . . . . . . . . . . . . . . 50 158 D.2. Since draft-ietf-httpbis-messaging-00 . . . . . . . . . . 51 159 D.3. Since draft-ietf-httpbis-messaging-01 . . . . . . . . . . 51 160 D.4. Since draft-ietf-httpbis-messaging-02 . . . . . . . . . . 52 161 D.5. Since draft-ietf-httpbis-messaging-03 . . . . . . . . . . 52 162 D.6. Since draft-ietf-httpbis-messaging-04 . . . . . . . . . . 52 163 D.7. Since draft-ietf-httpbis-messaging-05 . . . . . . . . . . 52 164 D.8. Since draft-ietf-httpbis-messaging-06 . . . . . . . . . . 53 165 D.9. Since draft-ietf-httpbis-messaging-07 . . . . . . . . . . 53 166 D.10. Since draft-ietf-httpbis-messaging-08 . . . . . . . . . . 54 167 D.11. Since draft-ietf-httpbis-messaging-09 . . . . . . . . . . 54 168 D.12. Since draft-ietf-httpbis-messaging-10 . . . . . . . . . . 54 169 D.13. Since draft-ietf-httpbis-messaging-11 . . . . . . . . . . 54 170 D.14. Since draft-ietf-httpbis-messaging-12 . . . . . . . . . . 55 171 D.15. Since draft-ietf-httpbis-messaging-13 . . . . . . . . . . 55 172 D.16. Since draft-ietf-httpbis-messaging-14 . . . . . . . . . . 55 173 D.17. Since draft-ietf-httpbis-messaging-15 . . . . . . . . . . 56 174 Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . 56 175 Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 176 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 59 178 1. Introduction 180 The Hypertext Transfer Protocol (HTTP) is a stateless application- 181 level request/response protocol that uses extensible semantics and 182 self-descriptive messages for flexible interaction with network-based 183 hypertext information systems. HTTP/1.1 is defined by: 185 * This document 187 * "HTTP Semantics" [Semantics] 189 * "HTTP Caching" [Caching] 190 This document specifies how HTTP semantics are conveyed using the 191 HTTP/1.1 message syntax, framing and connection management 192 mechanisms. Its goal is to define the complete set of requirements 193 for HTTP/1.1 message parsers and message-forwarding intermediaries. 195 This document obsoletes the portions of RFC 7230 related to HTTP/1.1 196 messaging and connection management, with the changes being 197 summarized in Appendix C.3. The other parts of RFC 7230 are 198 obsoleted by "HTTP Semantics" [Semantics]. 200 1.1. Requirements Notation 202 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 203 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 204 "OPTIONAL" in this document are to be interpreted as described in BCP 205 14 [RFC2119] [RFC8174] when, and only when, they appear in all 206 capitals, as shown here. 208 Conformance criteria and considerations regarding error handling are 209 defined in Section 2 of [Semantics]. 211 1.2. Syntax Notation 213 This specification uses the Augmented Backus-Naur Form (ABNF) 214 notation of [RFC5234], extended with the notation for case- 215 sensitivity in strings defined in [RFC7405]. 217 It also uses a list extension, defined in Section 5.6.1 of 218 [Semantics], that allows for compact definition of comma-separated 219 lists using a '#' operator (similar to how the '*' operator indicates 220 repetition). Appendix A shows the collected grammar with all list 221 operators expanded to standard ABNF notation. 223 As a convention, ABNF rule names prefixed with "obs-" denote 224 "obsolete" grammar rules that appear for historical reasons. 226 The following core rules are included by reference, as defined in 227 [RFC5234], Appendix B.1: ALPHA (letters), CR (carriage return), CRLF 228 (CR LF), CTL (controls), DIGIT (decimal 0-9), DQUOTE (double quote), 229 HEXDIG (hexadecimal 0-9/A-F/a-f), HTAB (horizontal tab), LF (line 230 feed), OCTET (any 8-bit sequence of data), SP (space), and VCHAR (any 231 visible [USASCII] character). 233 The rules below are defined in [Semantics]: 235 BWS = 236 OWS = 237 RWS = 238 absolute-path = 239 field-name = 240 field-value = 241 obs-text = 242 quoted-string = 243 token = 244 transfer-coding = 245 247 The rules below are defined in [RFC3986]: 249 absolute-URI = 250 authority = 251 uri-host = 252 port = 253 query = 255 2. Message 257 2.1. Message Format 259 An HTTP/1.1 message consists of a start-line followed by a CRLF and a 260 sequence of octets in a format similar to the Internet Message Format 261 [RFC5322]: zero or more header field lines (collectively referred to 262 as the "headers" or the "header section"), an empty line indicating 263 the end of the header section, and an optional message body. 265 HTTP-message = start-line CRLF 266 *( field-line CRLF ) 267 CRLF 268 [ message-body ] 270 A message can be either a request from client to server or a response 271 from server to client. Syntactically, the two types of message 272 differ only in the start-line, which is either a request-line (for 273 requests) or a status-line (for responses), and in the algorithm for 274 determining the length of the message body (Section 6). 276 start-line = request-line / status-line 278 In theory, a client could receive requests and a server could receive 279 responses, distinguishing them by their different start-line formats. 280 In practice, servers are implemented to only expect a request (a 281 response is interpreted as an unknown or invalid request method) and 282 clients are implemented to only expect a response. 284 Although HTTP makes use of some protocol elements similar to the 285 Multipurpose Internet Mail Extensions (MIME) [RFC2045], see 286 Appendix B for the differences between HTTP and MIME messages. 288 2.2. Message Parsing 290 The normal procedure for parsing an HTTP message is to read the 291 start-line into a structure, read each header field line into a hash 292 table by field name until the empty line, and then use the parsed 293 data to determine if a message body is expected. If a message body 294 has been indicated, then it is read as a stream until an amount of 295 octets equal to the message body length is read or the connection is 296 closed. 298 A recipient MUST parse an HTTP message as a sequence of octets in an 299 encoding that is a superset of US-ASCII [USASCII]. Parsing an HTTP 300 message as a stream of Unicode characters, without regard for the 301 specific encoding, creates security vulnerabilities due to the 302 varying ways that string processing libraries handle invalid 303 multibyte character sequences that contain the octet LF (%x0A). 304 String-based parsers can only be safely used within protocol elements 305 after the element has been extracted from the message, such as within 306 a header field line value after message parsing has delineated the 307 individual field lines. 309 Although the line terminator for the start-line and fields is the 310 sequence CRLF, a recipient MAY recognize a single LF as a line 311 terminator and ignore any preceding CR. 313 A sender MUST NOT generate a bare CR (a CR character not immediately 314 followed by LF) within any protocol elements other than the content. 315 A recipient of such a bare CR MUST consider that element to be 316 invalid or replace each bare CR with SP before processing the element 317 or forwarding the message. 319 Older HTTP/1.0 user agent implementations might send an extra CRLF 320 after a POST request as a workaround for some early server 321 applications that failed to read message body content that was not 322 terminated by a line-ending. An HTTP/1.1 user agent MUST NOT preface 323 or follow a request with an extra CRLF. If terminating the request 324 message body with a line-ending is desired, then the user agent MUST 325 count the terminating CRLF octets as part of the message body length. 327 In the interest of robustness, a server that is expecting to receive 328 and parse a request-line SHOULD ignore at least one empty line (CRLF) 329 received prior to the request-line. 331 A sender MUST NOT send whitespace between the start-line and the 332 first header field. A recipient that receives whitespace between the 333 start-line and the first header field MUST either reject the message 334 as invalid or consume each whitespace-preceded line without further 335 processing of it (i.e., ignore the entire line, along with any 336 subsequent lines preceded by whitespace, until a properly formed 337 header field is received or the header section is terminated). 339 The presence of such whitespace in a request might be an attempt to 340 trick a server into ignoring that field line or processing the line 341 after it as a new request, either of which might result in a security 342 vulnerability if other implementations within the request chain 343 interpret the same message differently. Likewise, the presence of 344 such whitespace in a response might be ignored by some clients or 345 cause others to cease parsing. 347 When a server listening only for HTTP request messages, or processing 348 what appears from the start-line to be an HTTP request message, 349 receives a sequence of octets that does not match the HTTP-message 350 grammar aside from the robustness exceptions listed above, the server 351 SHOULD respond with a 400 (Bad Request) response and close the 352 connection. 354 2.3. HTTP Version 356 HTTP uses a "." numbering scheme to indicate versions 357 of the protocol. This specification defines version "1.1". 358 Section 2.5 of [Semantics] specifies the semantics of HTTP version 359 numbers. 361 The version of an HTTP/1.x message is indicated by an HTTP-version 362 field in the start-line. HTTP-version is case-sensitive. 364 HTTP-version = HTTP-name "/" DIGIT "." DIGIT 365 HTTP-name = %s"HTTP" 367 When an HTTP/1.1 message is sent to an HTTP/1.0 recipient [RFC1945] 368 or a recipient whose version is unknown, the HTTP/1.1 message is 369 constructed such that it can be interpreted as a valid HTTP/1.0 370 message if all of the newer features are ignored. This specification 371 places recipient-version requirements on some new features so that a 372 conformant sender will only use compatible features until it has 373 determined, through configuration or the receipt of a message, that 374 the recipient supports HTTP/1.1. 376 Intermediaries that process HTTP messages (i.e., all intermediaries 377 other than those acting as tunnels) MUST send their own HTTP-version 378 in forwarded messages, unless it is purposefully downgraded as a 379 workaround for an upstream issue. In other words, an intermediary is 380 not allowed to blindly forward the start-line without ensuring that 381 the protocol version in that message matches a version to which that 382 intermediary is conformant for both the receiving and sending of 383 messages. Forwarding an HTTP message without rewriting the HTTP- 384 version might result in communication errors when downstream 385 recipients use the message sender's version to determine what 386 features are safe to use for later communication with that sender. 388 A server MAY send an HTTP/1.0 response to an HTTP/1.1 request if it 389 is known or suspected that the client incorrectly implements the HTTP 390 specification and is incapable of correctly processing later version 391 responses, such as when a client fails to parse the version number 392 correctly or when an intermediary is known to blindly forward the 393 HTTP-version even when it doesn't conform to the given minor version 394 of the protocol. Such protocol downgrades SHOULD NOT be performed 395 unless triggered by specific client attributes, such as when one or 396 more of the request header fields (e.g., User-Agent) uniquely match 397 the values sent by a client known to be in error. 399 3. Request Line 401 A request-line begins with a method token, followed by a single space 402 (SP), the request-target, another single space (SP), and ends with 403 the protocol version. 405 request-line = method SP request-target SP HTTP-version 407 Although the request-line grammar rule requires that each of the 408 component elements be separated by a single SP octet, recipients MAY 409 instead parse on whitespace-delimited word boundaries and, aside from 410 the CRLF terminator, treat any form of whitespace as the SP separator 411 while ignoring preceding or trailing whitespace; such whitespace 412 includes one or more of the following octets: SP, HTAB, VT (%x0B), FF 413 (%x0C), or bare CR. However, lenient parsing can result in request 414 smuggling security vulnerabilities if there are multiple recipients 415 of the message and each has its own unique interpretation of 416 robustness (see Section 11.2). 418 HTTP does not place a predefined limit on the length of a request- 419 line, as described in Section 2 of [Semantics]. A server that 420 receives a method longer than any that it implements SHOULD respond 421 with a 501 (Not Implemented) status code. A server that receives a 422 request-target longer than any URI it wishes to parse MUST respond 423 with a 414 (URI Too Long) status code (see Section 15.5.15 of 424 [Semantics]). 426 Various ad hoc limitations on request-line length are found in 427 practice. It is RECOMMENDED that all HTTP senders and recipients 428 support, at a minimum, request-line lengths of 8000 octets. 430 3.1. Method 432 The method token indicates the request method to be performed on the 433 target resource. The request method is case-sensitive. 435 method = token 437 The request methods defined by this specification can be found in 438 Section 9 of [Semantics], along with information regarding the HTTP 439 method registry and considerations for defining new methods. 441 3.2. Request Target 443 The request-target identifies the target resource upon which to apply 444 the request. The client derives a request-target from its desired 445 target URI. There are four distinct formats for the request-target, 446 depending on both the method being requested and whether the request 447 is to a proxy. 449 request-target = origin-form 450 / absolute-form 451 / authority-form 452 / asterisk-form 454 No whitespace is allowed in the request-target. Unfortunately, some 455 user agents fail to properly encode or exclude whitespace found in 456 hypertext references, resulting in those disallowed characters being 457 sent as the request-target in a malformed request-line. 459 Recipients of an invalid request-line SHOULD respond with either a 460 400 (Bad Request) error or a 301 (Moved Permanently) redirect with 461 the request-target properly encoded. A recipient SHOULD NOT attempt 462 to autocorrect and then process the request without a redirect, since 463 the invalid request-line might be deliberately crafted to bypass 464 security filters along the request chain. 466 A client MUST send a Host header field in all HTTP/1.1 request 467 messages. If the target URI includes an authority component, then a 468 client MUST send a field value for Host that is identical to that 469 authority component, excluding any userinfo subcomponent and its "@" 470 delimiter (Section 4.2.1 of [Semantics]). If the authority component 471 is missing or undefined for the target URI, then a client MUST send a 472 Host header field with an empty field value. 474 A server MUST respond with a 400 (Bad Request) status code to any 475 HTTP/1.1 request message that lacks a Host header field and to any 476 request message that contains more than one Host header field line or 477 a Host header field with an invalid field value. 479 3.2.1. origin-form 481 The most common form of request-target is the _origin-form_. 483 origin-form = absolute-path [ "?" query ] 485 When making a request directly to an origin server, other than a 486 CONNECT or server-wide OPTIONS request (as detailed below), a client 487 MUST send only the absolute path and query components of the target 488 URI as the request-target. If the target URI's path component is 489 empty, the client MUST send "/" as the path within the origin-form of 490 request-target. A Host header field is also sent, as defined in 491 Section 7.2 of [Semantics]. 493 For example, a client wishing to retrieve a representation of the 494 resource identified as 496 http://www.example.org/where?q=now 498 directly from the origin server would open (or reuse) a TCP 499 connection to port 80 of the host "www.example.org" and send the 500 lines: 502 GET /where?q=now HTTP/1.1 503 Host: www.example.org 505 followed by the remainder of the request message. 507 3.2.2. absolute-form 509 When making a request to a proxy, other than a CONNECT or server-wide 510 OPTIONS request (as detailed below), a client MUST send the target 511 URI in _absolute-form_ as the request-target. 513 absolute-form = absolute-URI 515 The proxy is requested to either service that request from a valid 516 cache, if possible, or make the same request on the client's behalf 517 to either the next inbound proxy server or directly to the origin 518 server indicated by the request-target. Requirements on such 519 "forwarding" of messages are defined in Section 7.6 of [Semantics]. 521 An example absolute-form of request-line would be: 523 GET http://www.example.org/pub/WWW/TheProject.html HTTP/1.1 525 A client MUST send a Host header field in an HTTP/1.1 request even if 526 the request-target is in the absolute-form, since this allows the 527 Host information to be forwarded through ancient HTTP/1.0 proxies 528 that might not have implemented Host. 530 When a proxy receives a request with an absolute-form of request- 531 target, the proxy MUST ignore the received Host header field (if any) 532 and instead replace it with the host information of the request- 533 target. A proxy that forwards such a request MUST generate a new 534 Host field value based on the received request-target rather than 535 forward the received Host field value. 537 When an origin server receives a request with an absolute-form of 538 request-target, the origin server MUST ignore the received Host 539 header field (if any) and instead use the host information of the 540 request-target. Note that if the request-target does not have an 541 authority component, an empty Host header field will be sent in this 542 case. 544 To allow for transition to the absolute-form for all requests in some 545 future version of HTTP, a server MUST accept the absolute-form in 546 requests, even though HTTP/1.1 clients will only send them in 547 requests to proxies. 549 3.2.3. authority-form 551 The _authority-form_ of request-target is only used for CONNECT 552 requests (Section 9.3.6 of [Semantics]). It consists of only the 553 uri-host and port number of the tunnel destination, separated by a 554 colon (":"). 556 authority-form = uri-host ":" port 558 When making a CONNECT request to establish a tunnel through one or 559 more proxies, a client MUST send only the host and port of the tunnel 560 destination as the request-target. The client obtains the host and 561 port from the target URI's authority component, except that it sends 562 the scheme's default port if the target URI elides the port. For 563 example, a CONNECT request to "http://www.example.com" looks like 565 CONNECT www.example.com:80 HTTP/1.1 566 Host: www.example.com 568 3.2.4. asterisk-form 570 The _asterisk-form_ of request-target is only used for a server-wide 571 OPTIONS request (Section 9.3.7 of [Semantics]). 573 asterisk-form = "*" 575 When a client wishes to request OPTIONS for the server as a whole, as 576 opposed to a specific named resource of that server, the client MUST 577 send only "*" (%x2A) as the request-target. For example, 579 OPTIONS * HTTP/1.1 581 If a proxy receives an OPTIONS request with an absolute-form of 582 request-target in which the URI has an empty path and no query 583 component, then the last proxy on the request chain MUST send a 584 request-target of "*" when it forwards the request to the indicated 585 origin server. 587 For example, the request 589 OPTIONS http://www.example.org:8001 HTTP/1.1 591 would be forwarded by the final proxy as 593 OPTIONS * HTTP/1.1 594 Host: www.example.org:8001 596 after connecting to port 8001 of host "www.example.org". 598 3.3. Reconstructing the Target URI 600 The target URI is the request-target when the request-target is in 601 absolute-form. In that case, a server will parse the URI into its 602 generic components for further evaluation. 604 Otherwise, the server reconstructs the target URI from the connection 605 context and various parts of the request message in order to identify 606 the target resource (Section 7.1 of [Semantics]): 608 * If the server's configuration provides for a fixed URI scheme, or 609 a scheme is provided by a trusted outbound gateway, that scheme is 610 used for the target URI. This is common in large-scale 611 deployments because a gateway server will receive the client's 612 connection context and replace that with their own connection to 613 the inbound server. Otherwise, if the request is received over a 614 secured connection, the target URI's scheme is "https"; if not, 615 the scheme is "http". 617 * If the request-target is in authority-form, the target URI's 618 authority component is the request-target. Otherwise, the target 619 URI's authority component is the field value of the Host header 620 field. If there is no Host header field or if its field value is 621 empty or invalid, the target URI's authority component is empty. 623 * If the request-target is in authority-form or asterisk-form, the 624 target URI's combined path and query component is empty. 625 Otherwise, the target URI's combined path and query component is 626 the request-target. 628 * The components of a reconstructed target URI, once determined as 629 above, can be recombined into absolute-URI form by concatenating 630 the scheme, "://", authority, and combined path and query 631 component. 633 Example 1: the following message received over a secure connection 635 GET /pub/WWW/TheProject.html HTTP/1.1 636 Host: www.example.org 638 has a target URI of 640 https://www.example.org/pub/WWW/TheProject.html 642 Example 2: the following message received over an insecure connection 644 OPTIONS * HTTP/1.1 645 Host: www.example.org:8080 647 has a target URI of 649 http://www.example.org:8080 651 If the target URI's authority component is empty and its URI scheme 652 requires a non-empty authority (as is the case for "http" and 653 "https"), the server can reject the request or determine whether a 654 configured default applies that is consistent with the incoming 655 connection's context. Context might include connection details like 656 address and port, what security has been applied, and locally-defined 657 information specific to that server's configuration. An empty 658 authority is replaced with the configured default before further 659 processing of the request. 661 Supplying a default name for authority within the context of a 662 secured connection is inherently unsafe if there is any chance that 663 the user agent's intended authority might differ from the selected 664 default. A server that can uniquely identify an authority from the 665 request context MAY use that identity as a default without this risk. 666 Alternatively, it might be better to redirect the request to a safe 667 resource that explains how to obtain a new client. 669 Note that reconstructing the client's target URI is only half of the 670 process for identifying a target resource. The other half is 671 determining whether that target URI identifies a resource for which 672 the server is willing and able to send a response, as defined in 673 Section 7.4 of [Semantics]. 675 4. Status Line 677 The first line of a response message is the status-line, consisting 678 of the protocol version, a space (SP), the status code, another 679 space, and ending with an OPTIONAL textual phrase describing the 680 status code. 682 status-line = HTTP-version SP status-code SP [reason-phrase] 684 Although the status-line grammar rule requires that each of the 685 component elements be separated by a single SP octet, recipients MAY 686 instead parse on whitespace-delimited word boundaries and, aside from 687 the line terminator, treat any form of whitespace as the SP separator 688 while ignoring preceding or trailing whitespace; such whitespace 689 includes one or more of the following octets: SP, HTAB, VT (%x0B), FF 690 (%x0C), or bare CR. However, lenient parsing can result in response 691 splitting security vulnerabilities if there are multiple recipients 692 of the message and each has its own unique interpretation of 693 robustness (see Section 11.1). 695 The status-code element is a 3-digit integer code describing the 696 result of the server's attempt to understand and satisfy the client's 697 corresponding request. The rest of the response message is to be 698 interpreted in light of the semantics defined for that status code. 699 See Section 15 of [Semantics] for information about the semantics of 700 status codes, including the classes of status code (indicated by the 701 first digit), the status codes defined by this specification, 702 considerations for the definition of new status codes, and the IANA 703 registry. 705 status-code = 3DIGIT 707 The reason-phrase element exists for the sole purpose of providing a 708 textual description associated with the numeric status code, mostly 709 out of deference to earlier Internet application protocols that were 710 more frequently used with interactive text clients. 712 reason-phrase = 1*( HTAB / SP / VCHAR / obs-text ) 714 A client SHOULD ignore the reason-phrase content because it is not a 715 reliable channel for information (it might be translated for a given 716 locale, overwritten by intermediaries, or discarded when the message 717 is forwarded via other versions of HTTP). A server MUST send the 718 space that separates status-code from the reason-phrase even when the 719 reason-phrase is absent (i.e., the status-line would end with the 720 three octets SP CR LF). 722 5. Field Syntax 724 Each field line consists of a case-insensitive field name followed by 725 a colon (":"), optional leading whitespace, the field line value, and 726 optional trailing whitespace. 728 field-line = field-name ":" OWS field-value OWS 730 Most HTTP field names and the rules for parsing within field values 731 are defined in Section 6.3 of [Semantics]. This section covers the 732 generic syntax for header field inclusion within, and extraction 733 from, HTTP/1.1 messages. 735 5.1. Field Line Parsing 737 Messages are parsed using a generic algorithm, independent of the 738 individual field names. The contents within a given field line value 739 are not parsed until a later stage of message interpretation (usually 740 after the message's entire field section has been processed). 742 No whitespace is allowed between the field name and colon. In the 743 past, differences in the handling of such whitespace have led to 744 security vulnerabilities in request routing and response handling. A 745 server MUST reject any received request message that contains 746 whitespace between a header field name and colon with a response 747 status code of 400 (Bad Request). A proxy MUST remove any such 748 whitespace from a response message before forwarding the message 749 downstream. 751 A field line value might be preceded and/or followed by optional 752 whitespace (OWS); a single SP preceding the field line value is 753 preferred for consistent readability by humans. The field line value 754 does not include any leading or trailing whitespace: OWS occurring 755 before the first non-whitespace octet of the field line value or 756 after the last non-whitespace octet of the field line value ought to 757 be excluded by parsers when extracting the field line value from a 758 field line. 760 5.2. Obsolete Line Folding 762 Historically, HTTP field line values could be extended over multiple 763 lines by preceding each extra line with at least one space or 764 horizontal tab (obs-fold). This specification deprecates such line 765 folding except within the message/http media type (Section 10.1). 767 obs-fold = OWS CRLF RWS 768 ; obsolete line folding 770 A sender MUST NOT generate a message that includes line folding 771 (i.e., that has any field line value that contains a match to the 772 obs-fold rule) unless the message is intended for packaging within 773 the message/http media type. 775 A server that receives an obs-fold in a request message that is not 776 within a message/http container MUST either reject the message by 777 sending a 400 (Bad Request), preferably with a representation 778 explaining that obsolete line folding is unacceptable, or replace 779 each received obs-fold with one or more SP octets prior to 780 interpreting the field value or forwarding the message downstream. 782 A proxy or gateway that receives an obs-fold in a response message 783 that is not within a message/http container MUST either discard the 784 message and replace it with a 502 (Bad Gateway) response, preferably 785 with a representation explaining that unacceptable line folding was 786 received, or replace each received obs-fold with one or more SP 787 octets prior to interpreting the field value or forwarding the 788 message downstream. 790 A user agent that receives an obs-fold in a response message that is 791 not within a message/http container MUST replace each received 792 obs-fold with one or more SP octets prior to interpreting the field 793 value. 795 6. Message Body 797 The message body (if any) of an HTTP/1.1 message is used to carry 798 content (Section 6.4 of [Semantics]) for the request or response. 799 The message body is identical to the content unless a transfer coding 800 has been applied, as described in Section 6.1. 802 message-body = *OCTET 804 The rules for determining when a message body is present in an 805 HTTP/1.1 message differ for requests and responses. 807 The presence of a message body in a request is signaled by a 808 Content-Length or Transfer-Encoding header field. Request message 809 framing is independent of method semantics. 811 The presence of a message body in a response depends on both the 812 request method to which it is responding and the response status code 813 (Section 4), and corresponds to when content is allowed; see 814 Section 6.4 of [Semantics]. 816 6.1. Transfer-Encoding 818 The Transfer-Encoding header field lists the transfer coding names 819 corresponding to the sequence of transfer codings that have been (or 820 will be) applied to the content in order to form the message body. 821 Transfer codings are defined in Section 7. 823 Transfer-Encoding = #transfer-coding 824 ; defined in [Semantics], Section 10.1.4 826 Transfer-Encoding is analogous to the Content-Transfer-Encoding field 827 of MIME, which was designed to enable safe transport of binary data 828 over a 7-bit transport service ([RFC2045], Section 6). However, safe 829 transport has a different focus for an 8bit-clean transfer protocol. 830 In HTTP's case, Transfer-Encoding is primarily intended to accurately 831 delimit dynamically generated content and to distinguish encodings 832 that are only applied for transport efficiency or security from those 833 that are characteristics of the selected resource. 835 A recipient MUST be able to parse the chunked transfer coding 836 (Section 7.1) because it plays a crucial role in framing messages 837 when the content size is not known in advance. A sender MUST NOT 838 apply chunked more than once to a message body (i.e., chunking an 839 already chunked message is not allowed). If any transfer coding 840 other than chunked is applied to a request's content, the sender MUST 841 apply chunked as the final transfer coding to ensure that the message 842 is properly framed. If any transfer coding other than chunked is 843 applied to a response's content, the sender MUST either apply chunked 844 as the final transfer coding or terminate the message by closing the 845 connection. 847 For example, 849 Transfer-Encoding: gzip, chunked 851 indicates that the content has been compressed using the gzip coding 852 and then chunked using the chunked coding while forming the message 853 body. 855 Unlike Content-Encoding (Section 8.4.1 of [Semantics]), Transfer- 856 Encoding is a property of the message, not of the representation, and 857 any recipient along the request/response chain MAY decode the 858 received transfer coding(s) or apply additional transfer coding(s) to 859 the message body, assuming that corresponding changes are made to the 860 Transfer-Encoding field value. Additional information about the 861 encoding parameters can be provided by other header fields not 862 defined by this specification. 864 Transfer-Encoding MAY be sent in a response to a HEAD request or in a 865 304 (Not Modified) response (Section 15.4.5 of [Semantics]) to a GET 866 request, neither of which includes a message body, to indicate that 867 the origin server would have applied a transfer coding to the message 868 body if the request had been an unconditional GET. This indication 869 is not required, however, because any recipient on the response chain 870 (including the origin server) can remove transfer codings when they 871 are not needed. 873 A server MUST NOT send a Transfer-Encoding header field in any 874 response with a status code of 1xx (Informational) or 204 (No 875 Content). A server MUST NOT send a Transfer-Encoding header field in 876 any 2xx (Successful) response to a CONNECT request (Section 9.3.6 of 877 [Semantics]). 879 Transfer-Encoding was added in HTTP/1.1. It is generally assumed 880 that implementations advertising only HTTP/1.0 support will not 881 understand how to process transfer-encoded content. A client MUST 882 NOT send a request containing Transfer-Encoding unless it knows the 883 server will handle HTTP/1.1 requests (or later minor revisions); such 884 knowledge might be in the form of specific user configuration or by 885 remembering the version of a prior received response. A server MUST 886 NOT send a response containing Transfer-Encoding unless the 887 corresponding request indicates HTTP/1.1 (or later minor revisions). 889 A server that receives a request message with a transfer coding it 890 does not understand SHOULD respond with 501 (Not Implemented). 892 6.2. Content-Length 894 When a message does not have a Transfer-Encoding header field, a 895 Content-Length header field (Section 8.6 of [Semantics]) can provide 896 the anticipated size, as a decimal number of octets, for potential 897 content. For messages that do include content, the Content-Length 898 field value provides the framing information necessary for 899 determining where the data (and message) ends. For messages that do 900 not include content, the Content-Length indicates the size of the 901 selected representation (Section 8.6 of [Semantics]). 903 A sender MUST NOT send a Content-Length header field in any message 904 that contains a Transfer-Encoding header field. 906 | *Note:* HTTP's use of Content-Length for message framing 907 | differs significantly from the same field's use in MIME, where 908 | it is an optional field used only within the "message/external- 909 | body" media-type. 911 6.3. Message Body Length 913 The length of a message body is determined by one of the following 914 (in order of precedence): 916 1. Any response to a HEAD request and any response with a 1xx 917 (Informational), 204 (No Content), or 304 (Not Modified) status 918 code is always terminated by the first empty line after the 919 header fields, regardless of the header fields present in the 920 message, and thus cannot contain a message body or trailer 921 section. 923 2. Any 2xx (Successful) response to a CONNECT request implies that 924 the connection will become a tunnel immediately after the empty 925 line that concludes the header fields. A client MUST ignore any 926 Content-Length or Transfer-Encoding header fields received in 927 such a message. 929 3. If a message is received with both a Transfer-Encoding and a 930 Content-Length header field, the Transfer-Encoding overrides the 931 Content-Length. Such a message might indicate an attempt to 932 perform request smuggling (Section 11.2) or response splitting 933 (Section 11.1) and ought to be handled as an error. An 934 intermediary that chooses to forward the message MUST first 935 remove the received Content-Length field and process the 936 Transfer-Encoding (as described below) prior to forwarding the 937 message downstream. 939 4. If a Transfer-Encoding header field is present and the chunked 940 transfer coding (Section 7.1) is the final encoding, the message 941 body length is determined by reading and decoding the chunked 942 data until the transfer coding indicates the data is complete. 944 If a Transfer-Encoding header field is present in a response and 945 the chunked transfer coding is not the final encoding, the 946 message body length is determined by reading the connection until 947 it is closed by the server. 949 If a Transfer-Encoding header field is present in a request and 950 the chunked transfer coding is not the final encoding, the 951 message body length cannot be determined reliably; the server 952 MUST respond with the 400 (Bad Request) status code and then 953 close the connection. 955 5. If a message is received without Transfer-Encoding and with an 956 invalid Content-Length header field, then the message framing is 957 invalid and the recipient MUST treat it as an unrecoverable 958 error, unless the field value can be successfully parsed as a 959 comma-separated list (Section 5.6.1 of [Semantics]), all values 960 in the list are valid, and all values in the list are the same. 961 If this is a request message, the server MUST respond with a 400 962 (Bad Request) status code and then close the connection. If this 963 is a response message received by a proxy, the proxy MUST close 964 the connection to the server, discard the received response, and 965 send a 502 (Bad Gateway) response to the client. If this is a 966 response message received by a user agent, the user agent MUST 967 close the connection to the server and discard the received 968 response. 970 6. If a valid Content-Length header field is present without 971 Transfer-Encoding, its decimal value defines the expected message 972 body length in octets. If the sender closes the connection or 973 the recipient times out before the indicated number of octets are 974 received, the recipient MUST consider the message to be 975 incomplete and close the connection. 977 7. If this is a request message and none of the above are true, then 978 the message body length is zero (no message body is present). 980 8. Otherwise, this is a response message without a declared message 981 body length, so the message body length is determined by the 982 number of octets received prior to the server closing the 983 connection. 985 Since there is no way to distinguish a successfully completed, close- 986 delimited response message from a partially received message 987 interrupted by network failure, a server SHOULD generate encoding or 988 length-delimited messages whenever possible. The close-delimiting 989 feature exists primarily for backwards compatibility with HTTP/1.0. 991 | *Note:* Request messages are never close-delimited because they 992 | are always explicitly framed by length or transfer coding, with 993 | the absence of both implying the request ends immediately after 994 | the header section. 996 A server MAY reject a request that contains a message body but not a 997 Content-Length by responding with 411 (Length Required). 999 Unless a transfer coding other than chunked has been applied, a 1000 client that sends a request containing a message body SHOULD use a 1001 valid Content-Length header field if the message body length is known 1002 in advance, rather than the chunked transfer coding, since some 1003 existing services respond to chunked with a 411 (Length Required) 1004 status code even though they understand the chunked transfer coding. 1005 This is typically because such services are implemented via a gateway 1006 that requires a content-length in advance of being called and the 1007 server is unable or unwilling to buffer the entire request before 1008 processing. 1010 A user agent that sends a request that contains a message body MUST 1011 send either a valid Content-Length header field or use the chunked 1012 transfer coding. A client MUST NOT use the chunked transfer encoding 1013 unless it knows the server will handle HTTP/1.1 (or later) requests; 1014 such knowledge can be in the form of specific user configuration or 1015 by remembering the version of a prior received response. 1017 If the final response to the last request on a connection has been 1018 completely received and there remains additional data to read, a user 1019 agent MAY discard the remaining data or attempt to determine if that 1020 data belongs as part of the prior message body, which might be the 1021 case if the prior message's Content-Length value is incorrect. A 1022 client MUST NOT process, cache, or forward such extra data as a 1023 separate response, since such behavior would be vulnerable to cache 1024 poisoning. 1026 7. Transfer Codings 1028 Transfer coding names are used to indicate an encoding transformation 1029 that has been, can be, or might need to be applied to a message's 1030 content in order to ensure "safe transport" through the network. 1031 This differs from a content coding in that the transfer coding is a 1032 property of the message rather than a property of the representation 1033 that is being transferred. 1035 All transfer-coding names are case-insensitive and ought to be 1036 registered within the HTTP Transfer Coding registry, as defined in 1037 Section 7.3. They are used in the Transfer-Encoding (Section 6.1) 1038 and TE (Section 10.1.4 of [Semantics]) header fields (the latter also 1039 defining the "transfer-coding" grammar). 1041 7.1. Chunked Transfer Coding 1043 The chunked transfer coding wraps content in order to transfer it as 1044 a series of chunks, each with its own size indicator, followed by an 1045 OPTIONAL trailer section containing trailer fields. Chunked enables 1046 content streams of unknown size to be transferred as a sequence of 1047 length-delimited buffers, which enables the sender to retain 1048 connection persistence and the recipient to know when it has received 1049 the entire message. 1051 chunked-body = *chunk 1052 last-chunk 1053 trailer-section 1054 CRLF 1056 chunk = chunk-size [ chunk-ext ] CRLF 1057 chunk-data CRLF 1058 chunk-size = 1*HEXDIG 1059 last-chunk = 1*("0") [ chunk-ext ] CRLF 1061 chunk-data = 1*OCTET ; a sequence of chunk-size octets 1063 The chunk-size field is a string of hex digits indicating the size of 1064 the chunk-data in octets. The chunked transfer coding is complete 1065 when a chunk with a chunk-size of zero is received, possibly followed 1066 by a trailer section, and finally terminated by an empty line. 1068 A recipient MUST be able to parse and decode the chunked transfer 1069 coding. 1071 HTTP/1.1 does not define any means to limit the size of a chunked 1072 response such that an intermediary can be assured of buffering the 1073 entire response. Additionally, very large chunk sizes may cause 1074 overflows or loss of precision if their values are not represented 1075 accurately in a receiving implementation. Therefore, recipients MUST 1076 anticipate potentially large decimal numerals and prevent parsing 1077 errors due to integer conversion overflows or precision loss due to 1078 integer representation. 1080 The chunked encoding does not define any parameters. Their presence 1081 SHOULD be treated as an error. 1083 7.1.1. Chunk Extensions 1085 The chunked encoding allows each chunk to include zero or more chunk 1086 extensions, immediately following the chunk-size, for the sake of 1087 supplying per-chunk metadata (such as a signature or hash), mid- 1088 message control information, or randomization of message body size. 1090 chunk-ext = *( BWS ";" BWS chunk-ext-name 1091 [ BWS "=" BWS chunk-ext-val ] ) 1093 chunk-ext-name = token 1094 chunk-ext-val = token / quoted-string 1096 The chunked encoding is specific to each connection and is likely to 1097 be removed or recoded by each recipient (including intermediaries) 1098 before any higher-level application would have a chance to inspect 1099 the extensions. Hence, use of chunk extensions is generally limited 1100 to specialized HTTP services such as "long polling" (where client and 1101 server can have shared expectations regarding the use of chunk 1102 extensions) or for padding within an end-to-end secured connection. 1104 A recipient MUST ignore unrecognized chunk extensions. A server 1105 ought to limit the total length of chunk extensions received in a 1106 request to an amount reasonable for the services provided, in the 1107 same way that it applies length limitations and timeouts for other 1108 parts of a message, and generate an appropriate 4xx (Client Error) 1109 response if that amount is exceeded. 1111 7.1.2. Chunked Trailer Section 1113 A trailer section allows the sender to include additional fields at 1114 the end of a chunked message in order to supply metadata that might 1115 be dynamically generated while the content is sent, such as a message 1116 integrity check, digital signature, or post-processing status. The 1117 proper use and limitations of trailer fields are defined in 1118 Section 6.5 of [Semantics]. 1120 trailer-section = *( field-line CRLF ) 1122 A recipient that decodes and removes the chunked encoding from a 1123 message (e.g., for storage or forwarding to a non-HTTP/1.1 peer) MUST 1124 discard any received trailer fields, store/forward them separately 1125 from the header fields, or selectively merge into the header section 1126 only those trailer fields corresponding to header field definitions 1127 that are understood by the recipient to explicitly permit and define 1128 how their corresponding trailer field value can be safely merged. 1130 7.1.3. Decoding Chunked 1132 A process for decoding the chunked transfer coding can be represented 1133 in pseudo-code as: 1135 length := 0 1136 read chunk-size, chunk-ext (if any), and CRLF 1137 while (chunk-size > 0) { 1138 read chunk-data and CRLF 1139 append chunk-data to content 1140 length := length + chunk-size 1141 read chunk-size, chunk-ext (if any), and CRLF 1142 } 1143 read trailer field 1144 while (trailer field is not empty) { 1145 if (trailer fields are stored/forwarded separately) { 1146 append trailer field to existing trailer fields 1147 } 1148 else if (trailer field is understood and defined as mergeable) { 1149 merge trailer field with existing header fields 1150 } 1151 else { 1152 discard trailer field 1153 } 1154 read trailer field 1155 } 1156 Content-Length := length 1157 Remove "chunked" from Transfer-Encoding 1159 7.2. Transfer Codings for Compression 1161 The following transfer coding names for compression are defined by 1162 the same algorithm as their corresponding content coding: 1164 compress (and x-compress) 1165 See Section 8.4.1.1 of [Semantics]. 1167 deflate 1168 See Section 8.4.1.2 of [Semantics]. 1170 gzip (and x-gzip) 1171 See Section 8.4.1.3 of [Semantics]. 1173 The compression codings do not define any parameters. Their presence 1174 SHOULD be treated as an error. 1176 7.3. Transfer Coding Registry 1178 The "HTTP Transfer Coding Registry" defines the namespace for 1179 transfer coding names. It is maintained at 1180 . 1182 Registrations MUST include the following fields: 1184 * Name 1186 * Description 1188 * Pointer to specification text 1190 Names of transfer codings MUST NOT overlap with names of content 1191 codings (Section 8.4.1 of [Semantics]) unless the encoding 1192 transformation is identical, as is the case for the compression 1193 codings defined in Section 7.2. 1195 The TE header field (Section 10.1.4 of [Semantics]) uses a pseudo 1196 parameter named "q" as rank value when multiple transfer codings are 1197 acceptable. Future registrations of transfer codings SHOULD NOT 1198 define parameters called "q" (case-insensitively) in order to avoid 1199 ambiguities. 1201 Values to be added to this namespace require IETF Review (see 1202 Section 4.8 of [RFC8126]), and MUST conform to the purpose of 1203 transfer coding defined in this specification. 1205 Use of program names for the identification of encoding formats is 1206 not desirable and is discouraged for future encodings. 1208 7.4. Negotiating Transfer Codings 1210 The TE field (Section 10.1.4 of [Semantics]) is used in HTTP/1.1 to 1211 indicate what transfer-codings, besides chunked, the client is 1212 willing to accept in the response, and whether or not the client is 1213 willing to preserve trailer fields in a chunked transfer coding. 1215 A client MUST NOT send the chunked transfer coding name in TE; 1216 chunked is always acceptable for HTTP/1.1 recipients. 1218 Three examples of TE use are below. 1220 TE: deflate 1221 TE: 1222 TE: trailers, deflate;q=0.5 1224 When multiple transfer codings are acceptable, the client MAY rank 1225 the codings by preference using a case-insensitive "q" parameter 1226 (similar to the qvalues used in content negotiation fields, 1227 Section 12.4.2 of [Semantics]). The rank value is a real number in 1228 the range 0 through 1, where 0.001 is the least preferred and 1 is 1229 the most preferred; a value of 0 means "not acceptable". 1231 If the TE field value is empty or if no TE field is present, the only 1232 acceptable transfer coding is chunked. A message with no transfer 1233 coding is always acceptable. 1235 The keyword "trailers" indicates that the sender will not discard 1236 trailer fields, as described in Section 6.5 of [Semantics]. 1238 Since the TE header field only applies to the immediate connection, a 1239 sender of TE MUST also send a "TE" connection option within the 1240 Connection header field (Section 7.6.1 of [Semantics]) in order to 1241 prevent the TE header field from being forwarded by intermediaries 1242 that do not support its semantics. 1244 8. Handling Incomplete Messages 1246 A server that receives an incomplete request message, usually due to 1247 a canceled request or a triggered timeout exception, MAY send an 1248 error response prior to closing the connection. 1250 A client that receives an incomplete response message, which can 1251 occur when a connection is closed prematurely or when decoding a 1252 supposedly chunked transfer coding fails, MUST record the message as 1253 incomplete. Cache requirements for incomplete responses are defined 1254 in Section 3 of [Caching]. 1256 If a response terminates in the middle of the header section (before 1257 the empty line is received) and the status code might rely on header 1258 fields to convey the full meaning of the response, then the client 1259 cannot assume that meaning has been conveyed; the client might need 1260 to repeat the request in order to determine what action to take next. 1262 A message body that uses the chunked transfer coding is incomplete if 1263 the zero-sized chunk that terminates the encoding has not been 1264 received. A message that uses a valid Content-Length is incomplete 1265 if the size of the message body received (in octets) is less than the 1266 value given by Content-Length. A response that has neither chunked 1267 transfer coding nor Content-Length is terminated by closure of the 1268 connection and, if the header section was received intact, is 1269 considered complete unless an error was indicated by the underlying 1270 connection (e.g., an "incomplete close" in TLS would leave the 1271 response incomplete, as described in Section 9.8). 1273 9. Connection Management 1275 HTTP messaging is independent of the underlying transport- or 1276 session-layer connection protocol(s). HTTP only presumes a reliable 1277 transport with in-order delivery of requests and the corresponding 1278 in-order delivery of responses. The mapping of HTTP request and 1279 response structures onto the data units of an underlying transport 1280 protocol is outside the scope of this specification. 1282 As described in Section 7.3 of [Semantics], the specific connection 1283 protocols to be used for an HTTP interaction are determined by client 1284 configuration and the target URI. For example, the "http" URI scheme 1285 (Section 4.2.1 of [Semantics]) indicates a default connection of TCP 1286 over IP, with a default TCP port of 80, but the client might be 1287 configured to use a proxy via some other connection, port, or 1288 protocol. 1290 HTTP implementations are expected to engage in connection management, 1291 which includes maintaining the state of current connections, 1292 establishing a new connection or reusing an existing connection, 1293 processing messages received on a connection, detecting connection 1294 failures, and closing each connection. Most clients maintain 1295 multiple connections in parallel, including more than one connection 1296 per server endpoint. Most servers are designed to maintain thousands 1297 of concurrent connections, while controlling request queues to enable 1298 fair use and detect denial-of-service attacks. 1300 9.1. Establishment 1302 It is beyond the scope of this specification to describe how 1303 connections are established via various transport- or session-layer 1304 protocols. Each connection applies to only one transport link. 1306 9.2. Associating a Response to a Request 1308 HTTP/1.1 does not include a request identifier for associating a 1309 given request message with its corresponding one or more response 1310 messages. Hence, it relies on the order of response arrival to 1311 correspond exactly to the order in which requests are made on the 1312 same connection. More than one response message per request only 1313 occurs when one or more informational responses (1xx, see 1314 Section 15.2 of [Semantics]) precede a final response to the same 1315 request. 1317 A client that has more than one outstanding request on a connection 1318 MUST maintain a list of outstanding requests in the order sent and 1319 MUST associate each received response message on that connection to 1320 the highest ordered request that has not yet received a final (non- 1321 1xx) response. 1323 If an HTTP/1.1 client receives data on a connection that doesn't have 1324 any outstanding requests, it MUST NOT consider them to be a response 1325 to a not-yet-issued request; it SHOULD close the connection, since 1326 message delimitation is now ambiguous, unless the data consists only 1327 of one or more CRLF (which can be discarded, as per Section 2.2). 1329 9.3. Persistence 1331 HTTP/1.1 defaults to the use of _persistent connections_, allowing 1332 multiple requests and responses to be carried over a single 1333 connection. HTTP implementations SHOULD support persistent 1334 connections. 1336 A recipient determines whether a connection is persistent or not 1337 based on the most recently received message's protocol version and 1338 Connection header field (Section 7.6.1 of [Semantics]), if any: 1340 * If the close connection option is present (Section 9.6), the 1341 connection will not persist after the current response; else, 1343 * If the received protocol is HTTP/1.1 (or later), the connection 1344 will persist after the current response; else, 1346 * If the received protocol is HTTP/1.0, the "keep-alive" connection 1347 option is present, either the recipient is not a proxy or the 1348 message is a response, and the recipient wishes to honor the 1349 HTTP/1.0 "keep-alive" mechanism, the connection will persist after 1350 the current response; otherwise, 1352 * The connection will close after the current response. 1354 A client that does not support persistent connections MUST send the 1355 close connection option in every request message. 1357 A server that does not support persistent connections MUST send the 1358 close connection option in every response message that does not have 1359 a 1xx (Informational) status code. 1361 A client MAY send additional requests on a persistent connection 1362 until it sends or receives a close connection option or receives an 1363 HTTP/1.0 response without a "keep-alive" connection option. 1365 In order to remain persistent, all messages on a connection need to 1366 have a self-defined message length (i.e., one not defined by closure 1367 of the connection), as described in Section 6. A server MUST read 1368 the entire request message body or close the connection after sending 1369 its response, since otherwise the remaining data on a persistent 1370 connection would be misinterpreted as the next request. Likewise, a 1371 client MUST read the entire response message body if it intends to 1372 reuse the same connection for a subsequent request. 1374 A proxy server MUST NOT maintain a persistent connection with an 1375 HTTP/1.0 client (see Section 19.7.1 of [RFC2068] for information and 1376 discussion of the problems with the Keep-Alive header field 1377 implemented by many HTTP/1.0 clients). 1379 See Appendix C.2.2 for more information on backwards compatibility 1380 with HTTP/1.0 clients. 1382 9.3.1. Retrying Requests 1384 Connections can be closed at any time, with or without intention. 1385 Implementations ought to anticipate the need to recover from 1386 asynchronous close events. The conditions under which a client can 1387 automatically retry a sequence of outstanding requests are defined in 1388 Section 9.2.2 of [Semantics]. 1390 9.3.2. Pipelining 1392 A client that supports persistent connections MAY _pipeline_ its 1393 requests (i.e., send multiple requests without waiting for each 1394 response). A server MAY process a sequence of pipelined requests in 1395 parallel if they all have safe methods (Section 9.2.1 of 1396 [Semantics]), but it MUST send the corresponding responses in the 1397 same order that the requests were received. 1399 A client that pipelines requests SHOULD retry unanswered requests if 1400 the connection closes before it receives all of the corresponding 1401 responses. When retrying pipelined requests after a failed 1402 connection (a connection not explicitly closed by the server in its 1403 last complete response), a client MUST NOT pipeline immediately after 1404 connection establishment, since the first remaining request in the 1405 prior pipeline might have caused an error response that can be lost 1406 again if multiple requests are sent on a prematurely closed 1407 connection (see the TCP reset problem described in Section 9.6). 1409 Idempotent methods (Section 9.2.2 of [Semantics]) are significant to 1410 pipelining because they can be automatically retried after a 1411 connection failure. A user agent SHOULD NOT pipeline requests after 1412 a non-idempotent method, until the final response status code for 1413 that method has been received, unless the user agent has a means to 1414 detect and recover from partial failure conditions involving the 1415 pipelined sequence. 1417 An intermediary that receives pipelined requests MAY pipeline those 1418 requests when forwarding them inbound, since it can rely on the 1419 outbound user agent(s) to determine what requests can be safely 1420 pipelined. If the inbound connection fails before receiving a 1421 response, the pipelining intermediary MAY attempt to retry a sequence 1422 of requests that have yet to receive a response if the requests all 1423 have idempotent methods; otherwise, the pipelining intermediary 1424 SHOULD forward any received responses and then close the 1425 corresponding outbound connection(s) so that the outbound user 1426 agent(s) can recover accordingly. 1428 9.4. Concurrency 1430 A client ought to limit the number of simultaneous open connections 1431 that it maintains to a given server. 1433 Previous revisions of HTTP gave a specific number of connections as a 1434 ceiling, but this was found to be impractical for many applications. 1435 As a result, this specification does not mandate a particular maximum 1436 number of connections but, instead, encourages clients to be 1437 conservative when opening multiple connections. 1439 Multiple connections are typically used to avoid the "head-of-line 1440 blocking" problem, wherein a request that takes significant server- 1441 side processing and/or transfers very large content would block 1442 subsequent requests on the same connection. However, each connection 1443 consumes server resources. Furthermore, using multiple connections 1444 can cause undesirable side effects in congested networks. 1446 Note that a server might reject traffic that it deems abusive or 1447 characteristic of a denial-of-service attack, such as an excessive 1448 number of open connections from a single client. 1450 9.5. Failures and Timeouts 1452 Servers will usually have some timeout value beyond which they will 1453 no longer maintain an inactive connection. Proxy servers might make 1454 this a higher value since it is likely that the client will be making 1455 more connections through the same proxy server. The use of 1456 persistent connections places no requirements on the length (or 1457 existence) of this timeout for either the client or the server. 1459 A client or server that wishes to time out SHOULD issue a graceful 1460 close on the connection. Implementations SHOULD constantly monitor 1461 open connections for a received closure signal and respond to it as 1462 appropriate, since prompt closure of both sides of a connection 1463 enables allocated system resources to be reclaimed. 1465 A client, server, or proxy MAY close the transport connection at any 1466 time. For example, a client might have started to send a new request 1467 at the same time that the server has decided to close the "idle" 1468 connection. From the server's point of view, the connection is being 1469 closed while it was idle, but from the client's point of view, a 1470 request is in progress. 1472 A server SHOULD sustain persistent connections, when possible, and 1473 allow the underlying transport's flow-control mechanisms to resolve 1474 temporary overloads, rather than terminate connections with the 1475 expectation that clients will retry. The latter technique can 1476 exacerbate network congestion or server load. 1478 A client sending a message body SHOULD monitor the network connection 1479 for an error response while it is transmitting the request. If the 1480 client sees a response that indicates the server does not wish to 1481 receive the message body and is closing the connection, the client 1482 SHOULD immediately cease transmitting the body and close its side of 1483 the connection. 1485 9.6. Tear-down 1487 The "close" connection option is defined as a signal that the sender 1488 will close this connection after completion of the response. A 1489 sender SHOULD send a Connection header field (Section 7.6.1 of 1490 [Semantics]) containing the close connection option when it intends 1491 to close a connection. For example, 1493 Connection: close 1495 as a request header field indicates that this is the last request 1496 that the client will send on this connection, while in a response the 1497 same field indicates that the server is going to close this 1498 connection after the response message is complete. 1500 Note that the field name "Close" is reserved, since using that name 1501 as a header field might conflict with the close connection option. 1503 A client that sends a close connection option MUST NOT send further 1504 requests on that connection (after the one containing the close) and 1505 MUST close the connection after reading the final response message 1506 corresponding to this request. 1508 A server that receives a close connection option MUST initiate 1509 closure of the connection (see below) after it sends the final 1510 response to the request that contained the close connection option. 1511 The server SHOULD send a close connection option in its final 1512 response on that connection. The server MUST NOT process any further 1513 requests received on that connection. 1515 A server that sends a close connection option MUST initiate closure 1516 of the connection (see below) after it sends the response containing 1517 the close connection option. The server MUST NOT process any further 1518 requests received on that connection. 1520 A client that receives a close connection option MUST cease sending 1521 requests on that connection and close the connection after reading 1522 the response message containing the close connection option; if 1523 additional pipelined requests had been sent on the connection, the 1524 client SHOULD NOT assume that they will be processed by the server. 1526 If a server performs an immediate close of a TCP connection, there is 1527 a significant risk that the client will not be able to read the last 1528 HTTP response. If the server receives additional data from the 1529 client on a fully closed connection, such as another request sent by 1530 the client before receiving the server's response, the server's TCP 1531 stack will send a reset packet to the client; unfortunately, the 1532 reset packet might erase the client's unacknowledged input buffers 1533 before they can be read and interpreted by the client's HTTP parser. 1535 To avoid the TCP reset problem, servers typically close a connection 1536 in stages. First, the server performs a half-close by closing only 1537 the write side of the read/write connection. The server then 1538 continues to read from the connection until it receives a 1539 corresponding close by the client, or until the server is reasonably 1540 certain that its own TCP stack has received the client's 1541 acknowledgement of the packet(s) containing the server's last 1542 response. Finally, the server fully closes the connection. 1544 It is unknown whether the reset problem is exclusive to TCP or might 1545 also be found in other transport connection protocols. 1547 Note that a TCP connection that is half-closed by the client does not 1548 delimit a request message, nor does it imply that the client is no 1549 longer interested in a response. In general, transport signals 1550 cannot be relied upon to signal edge cases, since HTTP/1.1 is 1551 independent of transport. 1553 9.7. TLS Connection Initiation 1555 Conceptually, HTTP/TLS is simply sending HTTP messages over a 1556 connection secured via TLS [RFC8446]. 1558 The HTTP client also acts as the TLS client. It initiates a 1559 connection to the server on the appropriate port and sends the TLS 1560 ClientHello to begin the TLS handshake. When the TLS handshake has 1561 finished, the client may then initiate the first HTTP request. All 1562 HTTP data MUST be sent as TLS "application data", but is otherwise 1563 treated like a normal connection for HTTP (including potential reuse 1564 as a persistent connection). 1566 9.8. TLS Connection Closure 1568 TLS provides a facility for secure connection closure. When a valid 1569 closure alert is received, an implementation can be assured that no 1570 further data will be received on that connection. TLS 1571 implementations MUST initiate an exchange of closure alerts before 1572 closing a connection. A TLS implementation MAY, after sending a 1573 closure alert, close the connection without waiting for the peer to 1574 send its closure alert, generating an "incomplete close". This 1575 SHOULD only be done when the application knows (typically through 1576 detecting HTTP message boundaries) that it has sent or received all 1577 the message data that it cares about. 1579 An incomplete close does not call into question the security of the 1580 data already received, but it could indicate that subsequent data 1581 might have been truncated. As TLS is not directly aware of HTTP 1582 message framing, it is necessary to examine the HTTP data itself to 1583 determine whether messages were complete. Handing of incomplete 1584 messages is defined in Section 8. 1586 When encountering an incomplete close, a client SHOULD treat as 1587 completed all requests for which it has received as much data as 1588 specified in the Content-Length header or, when a Transfer-Encoding 1589 of chunked is used, for which the terminal zero-length chunk has been 1590 received. A response that has neither chunked transfer coding nor 1591 Content-Length is complete only if a valid closure alert has been 1592 received. Treating an incomplete message as complete could expose 1593 implementations to attack. 1595 A client detecting an incomplete close SHOULD recover gracefully. 1597 Clients MUST send a closure alert before closing the connection. 1598 Clients that do not expect to receive any more data MAY choose not to 1599 wait for the server's closure alert and simply close the connection, 1600 thus generating an incomplete close on the server side. 1602 Servers SHOULD be prepared to receive an incomplete close from the 1603 client, since the client can often determine when the end of server 1604 data is. 1606 Servers MUST attempt to initiate an exchange of closure alerts with 1607 the client before closing the connection. Servers MAY close the 1608 connection after sending the closure alert, thus generating an 1609 incomplete close on the client side. 1611 10. Enclosing Messages as Data 1613 10.1. Media Type message/http 1615 The message/http media type can be used to enclose a single HTTP 1616 request or response message, provided that it obeys the MIME 1617 restrictions for all "message" types regarding line length and 1618 encodings. 1620 Type name: message 1622 Subtype name: http 1624 Required parameters: N/A 1626 Optional parameters: version, msgtype 1628 version: The HTTP-version number of the enclosed message (e.g., 1629 "1.1"). If not present, the version can be determined from the 1630 first line of the body. 1632 msgtype: The message type - "request" or "response". If not 1633 present, the type can be determined from the first line of the 1634 body. 1636 Encoding considerations: only "7bit", "8bit", or "binary" are 1637 permitted 1639 Security considerations: see Section 11 1641 Interoperability considerations: N/A 1643 Published specification: This specification (see Section 10.1). 1645 Applications that use this media type: N/A 1647 Fragment identifier considerations: N/A 1649 Additional information: Magic number(s): N/A 1650 Deprecated alias names for this type: N/A 1652 File extension(s): N/A 1654 Macintosh file type code(s): N/A 1656 Person and email address to contact for further information: See Aut 1657 hors' Addresses section. 1659 Intended usage: COMMON 1661 Restrictions on usage: N/A 1663 Author: See Authors' Addresses section. 1665 Change controller: IESG 1667 10.2. Media Type application/http 1669 The application/http media type can be used to enclose a pipeline of 1670 one or more HTTP request or response messages (not intermixed). 1672 Type name: application 1674 Subtype name: http 1676 Required parameters: N/A 1678 Optional parameters: version, msgtype 1680 version: The HTTP-version number of the enclosed messages (e.g., 1681 "1.1"). If not present, the version can be determined from the 1682 first line of the body. 1684 msgtype: The message type - "request" or "response". If not 1685 present, the type can be determined from the first line of the 1686 body. 1688 Encoding considerations: HTTP messages enclosed by this type are in 1689 "binary" format; use of an appropriate Content-Transfer-Encoding 1690 is required when transmitted via email. 1692 Security considerations: see Section 11 1694 Interoperability considerations: N/A 1696 Published specification: This specification (see Section 10.2). 1698 Applications that use this media type: N/A 1700 Fragment identifier considerations: N/A 1702 Additional information: Deprecated alias names for this type: N/A 1704 Magic number(s): N/A 1706 File extension(s): N/A 1708 Macintosh file type code(s): N/A 1710 Person and email address to contact for further information: See Aut 1711 hors' Addresses section. 1713 Intended usage: COMMON 1715 Restrictions on usage: N/A 1717 Author: See Authors' Addresses section. 1719 Change controller: IESG 1721 11. Security Considerations 1723 This section is meant to inform developers, information providers, 1724 and users of known security considerations relevant to HTTP message 1725 syntax and parsing. Security considerations about HTTP semantics, 1726 content, and routing are addressed in [Semantics]. 1728 11.1. Response Splitting 1730 Response splitting (a.k.a, CRLF injection) is a common technique, 1731 used in various attacks on Web usage, that exploits the line-based 1732 nature of HTTP message framing and the ordered association of 1733 requests to responses on persistent connections [Klein]. This 1734 technique can be particularly damaging when the requests pass through 1735 a shared cache. 1737 Response splitting exploits a vulnerability in servers (usually 1738 within an application server) where an attacker can send encoded data 1739 within some parameter of the request that is later decoded and echoed 1740 within any of the response header fields of the response. If the 1741 decoded data is crafted to look like the response has ended and a 1742 subsequent response has begun, the response has been split and the 1743 content within the apparent second response is controlled by the 1744 attacker. The attacker can then make any other request on the same 1745 persistent connection and trick the recipients (including 1746 intermediaries) into believing that the second half of the split is 1747 an authoritative answer to the second request. 1749 For example, a parameter within the request-target might be read by 1750 an application server and reused within a redirect, resulting in the 1751 same parameter being echoed in the Location header field of the 1752 response. If the parameter is decoded by the application and not 1753 properly encoded when placed in the response field, the attacker can 1754 send encoded CRLF octets and other content that will make the 1755 application's single response look like two or more responses. 1757 A common defense against response splitting is to filter requests for 1758 data that looks like encoded CR and LF (e.g., "%0D" and "%0A"). 1759 However, that assumes the application server is only performing URI 1760 decoding, rather than more obscure data transformations like charset 1761 transcoding, XML entity translation, base64 decoding, sprintf 1762 reformatting, etc. A more effective mitigation is to prevent 1763 anything other than the server's core protocol libraries from sending 1764 a CR or LF within the header section, which means restricting the 1765 output of header fields to APIs that filter for bad octets and not 1766 allowing application servers to write directly to the protocol 1767 stream. 1769 11.2. Request Smuggling 1771 Request smuggling ([Linhart]) is a technique that exploits 1772 differences in protocol parsing among various recipients to hide 1773 additional requests (which might otherwise be blocked or disabled by 1774 policy) within an apparently harmless request. Like response 1775 splitting, request smuggling can lead to a variety of attacks on HTTP 1776 usage. 1778 This specification has introduced new requirements on request 1779 parsing, particularly with regard to message framing in Section 6.3, 1780 to reduce the effectiveness of request smuggling. 1782 11.3. Message Integrity 1784 HTTP does not define a specific mechanism for ensuring message 1785 integrity, instead relying on the error-detection ability of 1786 underlying transport protocols and the use of length or chunk- 1787 delimited framing to detect completeness. Historically, the lack of 1788 a single integrity mechanism has been justified by the informal 1789 nature of most HTTP communication. However, the prevalence of HTTP 1790 as an information access mechanism has resulted in its increasing use 1791 within environments where verification of message integrity is 1792 crucial. 1794 The mechanisms provided with the "https" scheme, such as 1795 authenticated encryption, provide protection against modification of 1796 messages. Care is needed however to ensure that connection closure 1797 cannot be used to truncate messages (see Section 9.8). User agents 1798 might refuse to accept incomplete messages or treat them specially. 1799 For example, a browser being used to view medical history or drug 1800 interaction information needs to indicate to the user when such 1801 information is detected by the protocol to be incomplete, expired, or 1802 corrupted during transfer. Such mechanisms might be selectively 1803 enabled via user agent extensions or the presence of message 1804 integrity metadata in a response. 1806 The "http" scheme provides no protection against accidental or 1807 malicious modification of messages. 1809 Extensions to the protocol might be used to mitigate the risk of 1810 unwanted modification of messages by intermediaries, even when the 1811 "https" scheme is used. Integrity might be assured by using hash 1812 functions or digital signatures that are selectively added to 1813 messages via extensible metadata fields. 1815 11.4. Message Confidentiality 1817 HTTP relies on underlying transport protocols to provide message 1818 confidentiality when that is desired. HTTP has been specifically 1819 designed to be independent of the transport protocol, such that it 1820 can be used over many different forms of encrypted connection, with 1821 the selection of such transports being identified by the choice of 1822 URI scheme or within user agent configuration. 1824 The "https" scheme can be used to identify resources that require a 1825 confidential connection, as described in Section 4.2.2 of 1826 [Semantics]. 1828 12. IANA Considerations 1830 The change controller for the following registrations is: "IETF 1831 (iesg@ietf.org) - Internet Engineering Task Force". 1833 12.1. Field Name Registration 1835 First, introduce the new "Hypertext Transfer Protocol (HTTP) Field 1836 Name Registry" at as 1837 described in Section 18.4 of [Semantics]. 1839 Then, please update the registry with the field names listed in the 1840 table below: 1842 +===================+==========+======+============+ 1843 | Field Name | Status | Ref. | Comments | 1844 +===================+==========+======+============+ 1845 | Close | standard | 9.6 | (reserved) | 1846 +-------------------+----------+------+------------+ 1847 | MIME-Version | standard | B.1 | | 1848 +-------------------+----------+------+------------+ 1849 | Transfer-Encoding | standard | 6.1 | | 1850 +-------------------+----------+------+------------+ 1852 Table 1 1854 12.2. Media Type Registration 1856 Please update the "Media Types" registry at 1857 with the registration 1858 information in Section 10.1 and Section 10.2 for the media types 1859 "message/http" and "application/http", respectively. 1861 12.3. Transfer Coding Registration 1863 Please update the "HTTP Transfer Coding Registry" at 1864 with the 1865 registration procedure of Section 7.3 and the content coding names 1866 summarized in the table below. 1868 +============+===============================+===========+ 1869 | Name | Description | Reference | 1870 +============+===============================+===========+ 1871 | chunked | Transfer in a series of | Section | 1872 | | chunks | 7.1 | 1873 +------------+-------------------------------+-----------+ 1874 | compress | UNIX "compress" data format | Section | 1875 | | [Welch] | 7.2 | 1876 +------------+-------------------------------+-----------+ 1877 | deflate | "deflate" compressed data | Section | 1878 | | ([RFC1951]) inside the "zlib" | 7.2 | 1879 | | data format ([RFC1950]) | | 1880 +------------+-------------------------------+-----------+ 1881 | gzip | GZIP file format [RFC1952] | Section | 1882 | | | 7.2 | 1883 +------------+-------------------------------+-----------+ 1884 | trailers | (reserved) | Section | 1885 | | | 12.3 | 1886 +------------+-------------------------------+-----------+ 1887 | x-compress | Deprecated (alias for | Section | 1888 | | compress) | 7.2 | 1889 +------------+-------------------------------+-----------+ 1890 | x-gzip | Deprecated (alias for gzip) | Section | 1891 | | | 7.2 | 1892 +------------+-------------------------------+-----------+ 1894 Table 2 1896 | *Note:* the coding name "trailers" is reserved because its use 1897 | would conflict with the keyword "trailers" in the TE header 1898 | field (Section 10.1.4 of [Semantics]). 1900 12.4. ALPN Protocol ID Registration 1902 Please update the "TLS Application-Layer Protocol Negotiation (ALPN) 1903 Protocol IDs" registry at with the 1905 registration below: 1907 +==========+=============================+================+ 1908 | Protocol | Identification Sequence | Reference | 1909 +==========+=============================+================+ 1910 | HTTP/1.1 | 0x68 0x74 0x74 0x70 0x2f | (this | 1911 | | 0x31 0x2e 0x31 ("http/1.1") | specification) | 1912 +----------+-----------------------------+----------------+ 1914 Table 3 1916 13. References 1918 13.1. Normative References 1920 [Caching] Fielding, R., Ed., Nottingham, M., Ed., and J. Reschke, 1921 Ed., "HTTP Caching", Work in Progress, Internet-Draft, 1922 draft-ietf-httpbis-cache-16, 27 May 2021, 1923 . 1925 [RFC1950] Deutsch, L.P. and J-L. Gailly, "ZLIB Compressed Data 1926 Format Specification version 3.3", RFC 1950, 1927 DOI 10.17487/RFC1950, May 1996, 1928 . 1930 [RFC1951] Deutsch, P., "DEFLATE Compressed Data Format Specification 1931 version 1.3", RFC 1951, DOI 10.17487/RFC1951, May 1996, 1932 . 1934 [RFC1952] Deutsch, P., Gailly, J-L., Adler, M., Deutsch, L.P., and 1935 G. Randers-Pehrson, "GZIP file format specification 1936 version 4.3", RFC 1952, DOI 10.17487/RFC1952, May 1996, 1937 . 1939 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 1940 Requirement Levels", BCP 14, RFC 2119, 1941 DOI 10.17487/RFC2119, March 1997, 1942 . 1944 [RFC3986] Berners-Lee, T., Fielding, R., and L. Masinter, "Uniform 1945 Resource Identifier (URI): Generic Syntax", STD 66, 1946 RFC 3986, DOI 10.17487/RFC3986, January 2005, 1947 . 1949 [RFC5234] Crocker, D., Ed. and P. Overell, "Augmented BNF for Syntax 1950 Specifications: ABNF", STD 68, RFC 5234, 1951 DOI 10.17487/RFC5234, January 2008, 1952 . 1954 [RFC7405] Kyzivat, P., "Case-Sensitive String Support in ABNF", 1955 RFC 7405, DOI 10.17487/RFC7405, December 2014, 1956 . 1958 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 1959 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 1960 May 2017, . 1962 [RFC8446] Rescorla, E., "The Transport Layer Security (TLS) Protocol 1963 Version 1.3", RFC 8446, DOI 10.17487/RFC8446, August 2018, 1964 . 1966 [Semantics] 1967 Fielding, R., Ed., Nottingham, M., Ed., and J. Reschke, 1968 Ed., "HTTP Semantics", Work in Progress, Internet-Draft, 1969 draft-ietf-httpbis-semantics-16, 27 May 2021, 1970 . 1973 [USASCII] American National Standards Institute, "Coded Character 1974 Set -- 7-bit American Standard Code for Information 1975 Interchange", ANSI X3.4, 1986. 1977 [Welch] Welch, T. A., "A Technique for High-Performance Data 1978 Compression", IEEE Computer 17(6), June 1984. 1980 13.2. Informative References 1982 [Err4667] RFC Errata, Erratum ID 4667, RFC 7230, 1983 . 1985 [Klein] Klein, A., "Divide and Conquer - HTTP Response Splitting, 1986 Web Cache Poisoning Attacks, and Related Topics", March 1987 2004, . 1990 [Linhart] Linhart, C., Klein, A., Heled, R., and S. Orrin, "HTTP 1991 Request Smuggling", June 2005, 1992 . 1995 [RFC1945] Berners-Lee, T., Fielding, R.T., and H.F. Nielsen, 1996 "Hypertext Transfer Protocol -- HTTP/1.0", RFC 1945, 1997 DOI 10.17487/RFC1945, May 1996, 1998 . 2000 [RFC2045] Freed, N. and N.S. Borenstein, "Multipurpose Internet Mail 2001 Extensions (MIME) Part One: Format of Internet Message 2002 Bodies", RFC 2045, DOI 10.17487/RFC2045, November 1996, 2003 . 2005 [RFC2046] Freed, N. and N. Borenstein, "Multipurpose Internet Mail 2006 Extensions (MIME) Part Two: Media Types", RFC 2046, 2007 DOI 10.17487/RFC2046, November 1996, 2008 . 2010 [RFC2049] Freed, N. and N.S. Borenstein, "Multipurpose Internet Mail 2011 Extensions (MIME) Part Five: Conformance Criteria and 2012 Examples", RFC 2049, DOI 10.17487/RFC2049, November 1996, 2013 . 2015 [RFC2068] Fielding, R., Gettys, J., Mogul, J., Nielsen, H., and T. 2016 Berners-Lee, "Hypertext Transfer Protocol -- HTTP/1.1", 2017 RFC 2068, DOI 10.17487/RFC2068, January 1997, 2018 . 2020 [RFC2557] Palme, F., Hopmann, A., Shelness, N., and E. Stefferud, 2021 "MIME Encapsulation of Aggregate Documents, such as HTML 2022 (MHTML)", RFC 2557, DOI 10.17487/RFC2557, March 1999, 2023 . 2025 [RFC5322] Resnick, P., "Internet Message Format", RFC 5322, 2026 DOI 10.17487/RFC5322, October 2008, 2027 . 2029 [RFC7230] Fielding, R., Ed. and J. F. Reschke, Ed., "Hypertext 2030 Transfer Protocol (HTTP/1.1): Message Syntax and Routing", 2031 RFC 7230, DOI 10.17487/RFC7230, June 2014, 2032 . 2034 [RFC7231] Fielding, R., Ed. and J. F. Reschke, Ed., "Hypertext 2035 Transfer Protocol (HTTP/1.1): Semantics and Content", 2036 RFC 7231, DOI 10.17487/RFC7231, June 2014, 2037 . 2039 [RFC8126] Cotton, M., Leiba, B., and T. Narten, "Guidelines for 2040 Writing an IANA Considerations Section in RFCs", BCP 26, 2041 RFC 8126, DOI 10.17487/RFC8126, June 2017, 2042 . 2044 Appendix A. Collected ABNF 2046 In the collected ABNF below, list rules are expanded as per 2047 Section 5.6.1.1 of [Semantics]. 2049 BWS = 2051 HTTP-message = start-line CRLF *( field-line CRLF ) CRLF [ 2052 message-body ] 2053 HTTP-name = %x48.54.54.50 ; HTTP 2054 HTTP-version = HTTP-name "/" DIGIT "." DIGIT 2056 OWS = 2057 RWS = 2059 Transfer-Encoding = [ transfer-coding *( OWS "," OWS transfer-coding 2060 ) ] 2062 absolute-URI = 2063 absolute-form = absolute-URI 2064 absolute-path = 2065 asterisk-form = "*" 2066 authority = 2067 authority-form = uri-host ":" port 2069 chunk = chunk-size [ chunk-ext ] CRLF chunk-data CRLF 2070 chunk-data = 1*OCTET 2071 chunk-ext = *( BWS ";" BWS chunk-ext-name [ BWS "=" BWS chunk-ext-val 2072 ] ) 2073 chunk-ext-name = token 2074 chunk-ext-val = token / quoted-string 2075 chunk-size = 1*HEXDIG 2076 chunked-body = *chunk last-chunk trailer-section CRLF 2078 field-line = field-name ":" OWS field-value OWS 2079 field-name = 2080 field-value = 2082 last-chunk = 1*"0" [ chunk-ext ] CRLF 2084 message-body = *OCTET 2085 method = token 2087 obs-fold = OWS CRLF RWS 2088 obs-text = 2089 origin-form = absolute-path [ "?" query ] 2091 port = 2093 query = 2094 quoted-string = 2096 reason-phrase = 1*( HTAB / SP / VCHAR / obs-text ) 2097 request-line = method SP request-target SP HTTP-version 2098 request-target = origin-form / absolute-form / authority-form / 2099 asterisk-form 2101 start-line = request-line / status-line 2102 status-code = 3DIGIT 2103 status-line = HTTP-version SP status-code SP [ reason-phrase ] 2104 token = 2105 trailer-section = *( field-line CRLF ) 2106 transfer-coding = 2108 uri-host = 2110 Appendix B. Differences between HTTP and MIME 2112 HTTP/1.1 uses many of the constructs defined for the Internet Message 2113 Format [RFC5322] and the Multipurpose Internet Mail Extensions (MIME) 2114 [RFC2045] to allow a message body to be transmitted in an open 2115 variety of representations and with extensible fields. However, RFC 2116 2045 is focused only on email; applications of HTTP have many 2117 characteristics that differ from email; hence, HTTP has features that 2118 differ from MIME. These differences were carefully chosen to 2119 optimize performance over binary connections, to allow greater 2120 freedom in the use of new media types, to make date comparisons 2121 easier, and to acknowledge the practice of some early HTTP servers 2122 and clients. 2124 This appendix describes specific areas where HTTP differs from MIME. 2125 Proxies and gateways to and from strict MIME environments need to be 2126 aware of these differences and provide the appropriate conversions 2127 where necessary. 2129 B.1. MIME-Version 2131 HTTP is not a MIME-compliant protocol. However, messages can include 2132 a single MIME-Version header field to indicate what version of the 2133 MIME protocol was used to construct the message. Use of the MIME- 2134 Version header field indicates that the message is in full 2135 conformance with the MIME protocol (as defined in [RFC2045]). 2136 Senders are responsible for ensuring full conformance (where 2137 possible) when exporting HTTP messages to strict MIME environments. 2139 B.2. Conversion to Canonical Form 2141 MIME requires that an Internet mail body part be converted to 2142 canonical form prior to being transferred, as described in Section 4 2143 of [RFC2049], and that content with a type of "text" represent line 2144 breaks as CRLF, forbidding the use of CR or LF outside of line break 2145 sequences [RFC2046]. In contrast, HTTP does not care whether CRLF, 2146 bare CR, or bare LF are used to indicate a line break within content. 2148 A proxy or gateway from HTTP to a strict MIME environment ought to 2149 translate all line breaks within text media types to the RFC 2049 2150 canonical form of CRLF. Note, however, this might be complicated by 2151 the presence of a Content-Encoding and by the fact that HTTP allows 2152 the use of some charsets that do not use octets 13 and 10 to 2153 represent CR and LF, respectively. 2155 Conversion will break any cryptographic checksums applied to the 2156 original content unless the original content is already in canonical 2157 form. Therefore, the canonical form is recommended for any content 2158 that uses such checksums in HTTP. 2160 B.3. Conversion of Date Formats 2162 HTTP/1.1 uses a restricted set of date formats (Section 5.6.7 of 2163 [Semantics]) to simplify the process of date comparison. Proxies and 2164 gateways from other protocols ought to ensure that any Date header 2165 field present in a message conforms to one of the HTTP/1.1 formats 2166 and rewrite the date if necessary. 2168 B.4. Conversion of Content-Encoding 2170 MIME does not include any concept equivalent to HTTP/1.1's Content- 2171 Encoding header field. Since this acts as a modifier on the media 2172 type, proxies and gateways from HTTP to MIME-compliant protocols 2173 ought to either change the value of the Content-Type header field or 2174 decode the representation before forwarding the message. (Some 2175 experimental applications of Content-Type for Internet mail have used 2176 a media-type parameter of ";conversions=" to perform 2177 a function equivalent to Content-Encoding. However, this parameter 2178 is not part of the MIME standards). 2180 B.5. Conversion of Content-Transfer-Encoding 2182 HTTP does not use the Content-Transfer-Encoding field of MIME. 2183 Proxies and gateways from MIME-compliant protocols to HTTP need to 2184 remove any Content-Transfer-Encoding prior to delivering the response 2185 message to an HTTP client. 2187 Proxies and gateways from HTTP to MIME-compliant protocols are 2188 responsible for ensuring that the message is in the correct format 2189 and encoding for safe transport on that protocol, where "safe 2190 transport" is defined by the limitations of the protocol being used. 2191 Such a proxy or gateway ought to transform and label the data with an 2192 appropriate Content-Transfer-Encoding if doing so will improve the 2193 likelihood of safe transport over the destination protocol. 2195 B.6. MHTML and Line Length Limitations 2197 HTTP implementations that share code with MHTML [RFC2557] 2198 implementations need to be aware of MIME line length limitations. 2199 Since HTTP does not have this limitation, HTTP does not fold long 2200 lines. MHTML messages being transported by HTTP follow all 2201 conventions of MHTML, including line length limitations and folding, 2202 canonicalization, etc., since HTTP transfers message-bodies without 2203 modification and, aside from the "multipart/byteranges" type 2204 (Section 14.6 of [Semantics]), does not interpret the content or any 2205 MIME header lines that might be contained therein. 2207 Appendix C. Changes from previous RFCs 2209 C.1. Changes from HTTP/0.9 2211 Since HTTP/0.9 did not support header fields in a request, there is 2212 no mechanism for it to support name-based virtual hosts (selection of 2213 resource by inspection of the Host header field). Any server that 2214 implements name-based virtual hosts ought to disable support for 2215 HTTP/0.9. Most requests that appear to be HTTP/0.9 are, in fact, 2216 badly constructed HTTP/1.x requests caused by a client failing to 2217 properly encode the request-target. 2219 C.2. Changes from HTTP/1.0 2221 C.2.1. Multihomed Web Servers 2223 The requirements that clients and servers support the Host header 2224 field (Section 7.2 of [Semantics]), report an error if it is missing 2225 from an HTTP/1.1 request, and accept absolute URIs (Section 3.2) are 2226 among the most important changes defined by HTTP/1.1. 2228 Older HTTP/1.0 clients assumed a one-to-one relationship of IP 2229 addresses and servers; there was no other established mechanism for 2230 distinguishing the intended server of a request than the IP address 2231 to which that request was directed. The Host header field was 2232 introduced during the development of HTTP/1.1 and, though it was 2233 quickly implemented by most HTTP/1.0 browsers, additional 2234 requirements were placed on all HTTP/1.1 requests in order to ensure 2235 complete adoption. At the time of this writing, most HTTP-based 2236 services are dependent upon the Host header field for targeting 2237 requests. 2239 C.2.2. Keep-Alive Connections 2241 In HTTP/1.0, each connection is established by the client prior to 2242 the request and closed by the server after sending the response. 2243 However, some implementations implement the explicitly negotiated 2244 ("Keep-Alive") version of persistent connections described in 2245 Section 19.7.1 of [RFC2068]. 2247 Some clients and servers might wish to be compatible with these 2248 previous approaches to persistent connections, by explicitly 2249 negotiating for them with a "Connection: keep-alive" request header 2250 field. However, some experimental implementations of HTTP/1.0 2251 persistent connections are faulty; for example, if an HTTP/1.0 proxy 2252 server doesn't understand Connection, it will erroneously forward 2253 that header field to the next inbound server, which would result in a 2254 hung connection. 2256 One attempted solution was the introduction of a Proxy-Connection 2257 header field, targeted specifically at proxies. In practice, this 2258 was also unworkable, because proxies are often deployed in multiple 2259 layers, bringing about the same problem discussed above. 2261 As a result, clients are encouraged not to send the Proxy-Connection 2262 header field in any requests. 2264 Clients are also encouraged to consider the use of Connection: keep- 2265 alive in requests carefully; while they can enable persistent 2266 connections with HTTP/1.0 servers, clients using them will need to 2267 monitor the connection for "hung" requests (which indicate that the 2268 client ought stop sending the header field), and this mechanism ought 2269 not be used by clients at all when a proxy is being used. 2271 C.2.3. Introduction of Transfer-Encoding 2273 HTTP/1.1 introduces the Transfer-Encoding header field (Section 6.1). 2274 Transfer codings need to be decoded prior to forwarding an HTTP 2275 message over a MIME-compliant protocol. 2277 C.3. Changes from RFC 7230 2279 Most of the sections introducing HTTP's design goals, history, 2280 architecture, conformance criteria, protocol versioning, URIs, 2281 message routing, and header fields have been moved to [Semantics]. 2282 This document has been reduced to just the messaging syntax and 2283 connection management requirements specific to HTTP/1.1. 2285 Prohibited generation of bare CRs outside of content. (Section 2.2) 2286 The ABNF definition of authority-form has changed from the more 2287 general authority component of a URI (in which port is optional) to 2288 the specific host:port format that is required by CONNECT. 2289 (Section 3.2.3) 2291 In the ABNF for chunked extensions, re-introduced (bad) whitespace 2292 around ";" and "=". Whitespace was removed in [RFC7230], but that 2293 change was found to break existing implementations (see [Err4667]). 2294 (Section 7.1.1) 2296 Trailer field semantics now transcend the specifics of chunked 2297 encoding. The decoding algorithm for chunked (Section 7.1.3) has 2298 been updated to encourage storage/forwarding of trailer fields 2299 separately from the header section, to only allow merging into the 2300 header section if the recipient knows the corresponding field 2301 definition permits and defines how to merge, and otherwise to discard 2302 the trailer fields instead of merging. The trailer part is now 2303 called the trailer section to be more consistent with the header 2304 section and more distinct from a body part. (Section 7.1.2) 2306 Disallowed transfer coding parameters called "q" in order to avoid 2307 conflicts with the use of ranks in the TE header field. 2308 (Section 7.3) 2310 Appendix D. Change Log 2312 This section is to be removed before publishing as an RFC. 2314 D.1. Between RFC7230 and draft 00 2316 The changes were purely editorial: 2318 * Change boilerplate and abstract to indicate the "draft" status, 2319 and update references to ancestor specifications. 2321 * Adjust historical notes. 2323 * Update links to sibling specifications. 2325 * Replace sections listing changes from RFC 2616 by new empty 2326 sections referring to RFC 723x. 2328 * Remove acknowledgements specific to RFC 723x. 2330 * Move "Acknowledgements" to the very end and make them unnumbered. 2332 D.2. Since draft-ietf-httpbis-messaging-00 2334 The changes in this draft are editorial, with respect to HTTP as a 2335 whole, to move all core HTTP semantics into [Semantics]: 2337 * Moved introduction, architecture, conformance, and ABNF extensions 2338 from RFC 7230 (Messaging) to semantics [Semantics]. 2340 * Moved discussion of MIME differences from RFC 7231 (Semantics) to 2341 Appendix B since they mostly cover transforming 1.1 messages. 2343 * Moved all extensibility tips, registration procedures, and 2344 registry tables from the IANA considerations to normative 2345 sections, reducing the IANA considerations to just instructions 2346 that will be removed prior to publication as an RFC. 2348 D.3. Since draft-ietf-httpbis-messaging-01 2350 * Cite RFC 8126 instead of RFC 5226 () 2353 * Resolved erratum 4779, no change needed here 2354 (, 2355 ) 2357 * In Section 7, fixed prose claiming transfer parameters allow bare 2358 names (, 2359 ) 2361 * Resolved erratum 4225, no change needed here 2362 (, 2363 ) 2365 * Replace "response code" with "response status code" 2366 (, 2367 ) 2369 * In Section 9.3, clarify statement about HTTP/1.0 keep-alive 2370 (, 2371 ) 2373 * In Section 7.1.1, re-introduce (bad) whitespace around ";" and "=" 2374 (, 2375 , ) 2378 * In Section 7.3, state that transfer codings should not use 2379 parameters named "q" (, ) 2382 * In Section 7, mark coding name "trailers" as reserved in the IANA 2383 registry () 2385 D.4. Since draft-ietf-httpbis-messaging-02 2387 * In Section 4, explain why the reason phrase should be ignored by 2388 clients (). 2390 * Add Section 9.2 to explain how request/response correlation is 2391 performed () 2393 D.5. Since draft-ietf-httpbis-messaging-03 2395 * In Section 9.2, caution against treating data on a connection as 2396 part of a not-yet-issued request () 2399 * In Section 7, remove the predefined codings from the ABNF and make 2400 it generic instead () 2403 * Use RFC 7405 ABNF notation for case-sensitive string constants 2404 () 2406 D.6. Since draft-ietf-httpbis-messaging-04 2408 * In Section 7.8 of [Semantics], clarify that protocol-name is to be 2409 matched case-insensitively () 2412 * In Section 5.2, add leading optional whitespace to obs-fold ABNF 2413 (, 2414 ) 2416 * In Section 4, add clarifications about empty reason phrases 2417 () 2419 * Move discussion of retries from Section 9.3.1 into [Semantics] 2420 () 2422 D.7. Since draft-ietf-httpbis-messaging-05 2423 * In Section 7.1.2, the trailer part has been renamed the trailer 2424 section (for consistency with the header section) and trailers are 2425 no longer merged as header fields by default, but rather can be 2426 discarded, kept separate from header fields, or merged with header 2427 fields only if understood and defined as being mergeable 2428 () 2430 * In Section 2.1 and related Sections, move the trailing CRLF from 2431 the line grammars into the message format 2432 () 2434 * Moved Section 2.3 down () 2437 * In Section 7.8 of [Semantics], use 'websocket' instead of 2438 'HTTP/2.0' in examples () 2441 * Move version non-specific text from Section 6 into semantics as 2442 "payload" () 2444 * In Section 9.8, add text from RFC 2818 2445 () 2447 D.8. Since draft-ietf-httpbis-messaging-06 2449 * In Section 12.4, update the APLN protocol id for HTTP/1.1 2450 () 2452 * In Section 5, align with updates to field terminology in semantics 2453 () 2455 * In Section 7.6.1 of [Semantics], clarify that new connection 2456 options indeed need to be registered () 2459 * In Section 1.1, reference RFC 8174 as well 2460 () 2462 D.9. Since draft-ietf-httpbis-messaging-07 2464 * Move TE: trailers into [Semantics] () 2467 * In Section 6.3, adjust requirements for handling multiple content- 2468 length values () 2470 * Throughout, replace "effective request URI" with "target URI" 2471 () 2473 * In Section 6.1, don't claim Transfer-Encoding is supported by 2474 HTTP/2 or later () 2476 D.10. Since draft-ietf-httpbis-messaging-08 2478 * In Section 2.2, disallow bare CRs () 2481 * Appendix A now uses the sender variant of the "#" list expansion 2482 () 2484 * In Section 5, adjust IANA "Close" entry for new registry format 2485 () 2487 D.11. Since draft-ietf-httpbis-messaging-09 2489 * Switch to xml2rfc v3 mode for draft generation 2490 () 2492 D.12. Since draft-ietf-httpbis-messaging-10 2494 * In Section 6.3, note that TCP half-close does not delimit a 2495 request; talk about corresponding server-side behaviour in 2496 Section 9.6 () 2498 * Moved requirements specific to HTTP/1.1 from [Semantics] into 2499 Section 3.2 () 2501 * In Section 6.1 (Transfer-Encoding), adjust ABNF to allow empty 2502 lists () 2504 * In Section 9.7, add text from RFC 2818 2505 () 2507 * Moved definitions of "TE" and "Upgrade" into [Semantics] 2508 () 2510 * Moved definition of "Connection" into [Semantics] 2511 () 2513 D.13. Since draft-ietf-httpbis-messaging-11 2515 * Move IANA Upgrade Token Registry instructions to [Semantics] 2516 () 2518 D.14. Since draft-ietf-httpbis-messaging-12 2520 * Moved content of history appendix to Semantics 2521 () 2523 * Moved note about "close" being reserved as field name to 2524 Section 9.3 () 2526 * Moved table of transfer codings into Section 12.3 2527 () 2529 * In Section 13.2, updated the URI for the [Linhart] paper 2530 () 2532 * Changed document title to just "HTTP/1.1" 2533 () 2535 * In Section 7, moved transfer-coding ABNF to Section 10.1.4 of 2536 [Semantics] () 2538 * Changed to using "payload data" when defining requirements about 2539 the data being conveyed within a message, instead of the terms 2540 "payload body" or "response body" or "representation body", since 2541 they often get confused with the HTTP/1.1 message body (which 2542 includes transfer coding) () 2545 D.15. Since draft-ietf-httpbis-messaging-13 2547 * In Section 6.3, clarify that a message needs to be checked for 2548 both Content-Length and Transfer-Encoding, before processing 2549 Transfer-Encoding, and that ought to be treated as an error, but 2550 an intermediary can choose to forward the message downstream after 2551 removing the Content-Length and processing the Transfer-Encoding 2552 () 2554 * Changed to using "content" instead of "payload" or "payload data" 2555 to avoid confusion with the payload of version-specific messaging 2556 frames () 2558 D.16. Since draft-ietf-httpbis-messaging-14 2560 * In Section 9.6, define the close connection option, since its 2561 definition was removed from the Connection header field for being 2562 specific to 1.1 () 2564 * In Section 3.3, clarify how the target URI is reconstructed when 2565 the request-target is not in absolute-form and highlight risk in 2566 selecting a default host () 2569 * In Section 7.1, clarify large chunk handling issues 2570 () 2572 * In Section 2.2, explicitly close the connection after sending a 2573 400 () 2575 * In Section 2.3, refine version requirements for intermediaries 2576 () 2578 * In Section 7.1.3, don't remove the Trailer header field 2579 () 2581 * In Section 3.2.3, changed the ABNF definition of authority-form 2582 from the authority component (in which port is optional) to the 2583 host:port format that has always been required by CONNECT 2584 () 2586 D.17. Since draft-ietf-httpbis-messaging-15 2588 * None. 2590 Acknowledgments 2592 See Appendix "Acknowledgments" of [Semantics]. 2594 Index 2596 A C D F G H M O R T X 2598 A 2600 absolute-form (of request-target) Section 3.2.2 2601 application/http Media Type Section 10.2 2602 asterisk-form (of request-target) Section 3.2.4 2603 authority-form (of request-target) Section 3.2.3 2605 C 2607 Connection header field Section 9.6 2608 Content-Length header field Section 6.2 2609 Content-Transfer-Encoding header field Appendix B.5 2610 chunked (Coding Format) Section 6.1; Section 6.3 2611 chunked (transfer coding) Section 7.1 2612 close Section 9.3; Section 9.6 2613 compress (transfer coding) Section 7.2 2615 D 2617 deflate (transfer coding) Section 7.2 2619 F 2621 Fields 2622 Close Section 9.6, Paragraph 4 2623 MIME-Version Appendix B.1 2624 Transfer-Encoding Section 6.1 2626 G 2628 Grammar 2629 ALPHA Section 1.2 2630 CR Section 1.2 2631 CRLF Section 1.2 2632 CTL Section 1.2 2633 DIGIT Section 1.2 2634 DQUOTE Section 1.2 2635 HEXDIG Section 1.2 2636 HTAB Section 1.2 2637 HTTP-message Section 2.1 2638 HTTP-name Section 2.3 2639 HTTP-version Section 2.3 2640 LF Section 1.2 2641 OCTET Section 1.2 2642 SP Section 1.2 2643 Transfer-Encoding Section 6.1 2644 VCHAR Section 1.2 2645 absolute-form Section 3.2; Section 3.2.2 2646 asterisk-form Section 3.2; Section 3.2.4 2647 authority-form Section 3.2; Section 3.2.3 2648 chunk Section 7.1 2649 chunk-data Section 7.1 2650 chunk-ext Section 7.1; Section 7.1.1 2651 chunk-ext-name Section 7.1.1 2652 chunk-ext-val Section 7.1.1 2653 chunk-size Section 7.1 2654 chunked-body Section 7.1 2655 field-line Section 5; Section 7.1.2 2656 field-name Section 5 2657 field-value Section 5 2658 last-chunk Section 7.1 2659 message-body Section 6 2660 method Section 3.1 2661 obs-fold Section 5.2 2662 origin-form Section 3.2; Section 3.2.1 2663 reason-phrase Section 4 2664 request-line Section 3 2665 request-target Section 3.2 2666 start-line Section 2.1 2667 status-code Section 4 2668 status-line Section 4 2669 trailer-section Section 7.1; Section 7.1.2 2670 gzip (transfer coding) Section 7.2 2672 H 2674 Header Fields 2675 MIME-Version Appendix B.1 2676 Transfer-Encoding Section 6.1 2677 header line Section 2.1 2678 header section Section 2.1 2679 headers Section 2.1 2681 M 2683 MIME-Version header field Appendix B.1 2684 Media Type 2685 application/http Section 10.2 2686 message/http Section 10.1 2687 message/http Media Type Section 10.1 2688 method Section 3.1 2690 O 2692 origin-form (of request-target) Section 3.2.1 2694 R 2696 request-target Section 3.2 2698 T 2700 Transfer-Encoding header field Section 6.1 2702 X 2704 x-compress (transfer coding) Section 7.2 2705 x-gzip (transfer coding) Section 7.2 2707 Authors' Addresses 2709 Roy T. Fielding (editor) 2710 Adobe 2711 345 Park Ave 2712 San Jose, CA 95110 2713 United States of America 2715 Email: fielding@gbiv.com 2716 URI: https://roy.gbiv.com/ 2718 Mark Nottingham (editor) 2719 Fastly 2720 Prahran VIC 2721 Australia 2723 Email: mnot@mnot.net 2724 URI: https://www.mnot.net/ 2726 Julian Reschke (editor) 2727 greenbytes GmbH 2728 Hafenweg 16 2729 48155 Münster 2730 Germany 2732 Email: julian.reschke@greenbytes.de 2733 URI: https://greenbytes.de/tech/webdav/