rfc9112.xml

<?xml version="1.0" encoding="utf-8"?>
<?xml-stylesheet type='text/xsl' href='lib/myxml2rfc.xslt'?>
<!DOCTYPE rfc [
  <!ENTITY MAY "<bcp14>MAY</bcp14>">
  <!ENTITY MUST "<bcp14>MUST</bcp14>">
  <!ENTITY MUST-NOT "<bcp14>MUST NOT</bcp14>">
  <!ENTITY NOT-RECOMMENDED "<bcp14>NOT RECOMMENDED</bcp14>">
  <!ENTITY OPTIONAL "<bcp14>OPTIONAL</bcp14>">
  <!ENTITY RECOMMENDED "<bcp14>RECOMMENDED</bcp14>">
  <!ENTITY REQUIRED "<bcp14>REQUIRED</bcp14>">
  <!ENTITY SHALL "<bcp14>SHALL</bcp14>">
  <!ENTITY SHALL-NOT "<bcp14>SHALL NOT</bcp14>">
  <!ENTITY SHOULD "<bcp14>SHOULD</bcp14>">
  <!ENTITY SHOULD-NOT "<bcp14>SHOULD NOT</bcp14>">
  <!ENTITY ID-VERSION "latest">
  <!ENTITY mdash "&#8212;">
  <!ENTITY nbsp "&#160;">
  <!ENTITY uuml "&#252;">
  <!ENTITY Note "<x:h xmlns:x='http://purl.org/net/xml2rfc/ext'>Note:</x:h>">
]>
<?rfc compact="yes"?>
<?rfc subcompact="no" ?>
<?rfc linkmailto="no" ?>
<?rfc editing="no" ?>
<?rfc comments="yes"?>
<?rfc inline="yes"?>
<?rfc rfcedstyle="yes"?>
<?rfc-ext allow-markup-in-artwork="yes" ?>
<?rfc-ext map-dfn="none" ?>
<?rfc-ext html-pretty-print="prettyprint https://cdn.rawgit.com/google/code-prettify/master/loader/run_prettify.js"?>
<?rfc-ext include-references-in-index="yes" ?>
<?rfc-ext xml2rfc-backend="202007"?>

<rfc obsoletes="7230" category="std" x:maturity-level="internet"
     number="9112" seriesNo="99"
     ipr="pre5378Trust200902" docName="draft-ietf-httpbis-messaging-&ID-VERSION;"
     xmlns:x='http://purl.org/net/xml2rfc/ext'
     xmlns:rdf='http://www.w3.org/1999/02/22-rdf-syntax-ns#'
     version="3" tocDepth="4" sortRefs="true">
<x:feedback template="mailto:ietf-http-wg@w3.org?subject={docname},%20%22{section}%22&amp;body=&lt;{ref}&gt;:"/>
<front>

  <title>HTTP/1.1</title>

  <author fullname="Roy T. Fielding" initials="R." surname="Fielding" role="editor">
    <organization>Adobe</organization>
    <address>
      <postal>
        <postalLine>345 Park Ave</postalLine>
        <postalLine>San Jose, CA 95110</postalLine>
        <country>United States of America</country>
      </postal>
      <email>fielding@gbiv.com</email>
      <uri>https://roy.gbiv.com/</uri>
    </address>
  </author>

  <author fullname="Mark Nottingham" initials="M." surname="Nottingham" role="editor">
    <organization>Fastly</organization>
    <address>
       <postal>
         <postalLine>Prahran</postalLine>
         <country>Australia</country>
       </postal>
      <email>mnot@mnot.net</email>
      <uri>https://www.mnot.net/</uri>
    </address>
  </author>

  <author fullname="Julian Reschke" initials="J." surname="Reschke" role="editor">
    <organization abbrev="greenbytes">greenbytes GmbH</organization>
    <address>
      <postal>
        <postalLine>Hafenweg 16</postalLine>
        <postalLine>48155 M&uuml;nster</postalLine>
        <country>Germany</country>
      </postal>
      <email>julian.reschke@greenbytes.de</email>
      <uri>https://greenbytes.de/tech/webdav/</uri>
    </address>
  </author>

  <date year="2022" month="June"/>

  <area>Applications and Real-Time</area>
  <workgroup>HTTP</workgroup>

  <keyword>Hypertext Transfer Protocol</keyword>
  <keyword>HTTP</keyword>
  <keyword>HTTP message format</keyword>

<abstract>
<t>
   The Hypertext Transfer Protocol (HTTP) is a stateless application-level
   protocol for distributed, collaborative, hypertext information systems.
   This document specifies the HTTP/1.1 message syntax, message parsing,
   connection management, and related security concerns.
</t>
<t>
   This document obsoletes portions of RFC 7230.
</t>
</abstract>
</front>
<middle>
<section title="Introduction" anchor="introduction">
<t>
   The Hypertext Transfer Protocol (HTTP) is a stateless application-level
   request/response protocol that uses extensible semantics and
   self-descriptive messages for flexible interaction with network-based
   hypertext information systems. HTTP/1.1 is defined by:
</t>
<ul>
   <li>This document</li>
   <li>"HTTP Semantics" <xref target="HTTP"/></li>
   <li>"HTTP Caching" <xref target="CACHING"/></li>
</ul>
<t>
   This document specifies how HTTP semantics are conveyed using the
   HTTP/1.1 message syntax, framing, and connection management mechanisms.
   Its goal is to define the complete set of requirements for HTTP/1.1
   message parsers and message-forwarding intermediaries.
</t>
<t>
   This document obsoletes the portions of
   <xref target="RFC7230" format="none">RFC 7230</xref> related to HTTP/1.1
   messaging and connection management, with the changes being summarized in
   <xref target="changes.from.rfc.7230"/>. The other parts of
   <xref target="RFC7230" format="none">RFC 7230</xref> are obsoleted by
   "HTTP Semantics" <xref target="HTTP"/>.
</t>

<section title="Requirements Notation" anchor="requirements.notation">
<t>
   The key words "&MUST;", "&MUST-NOT;", "&REQUIRED;", "&SHALL;", "&SHALL-NOT;",
   "&SHOULD;", "&SHOULD-NOT;", "&RECOMMENDED;", "&NOT-RECOMMENDED;",
   "&MAY;", and "&OPTIONAL;" in this document are to be interpreted as
   described in BCP 14 <xref target="RFC2119"/> <xref target="RFC8174"/> when, and only when, they
   appear in all capitals, as shown here.
</t>
<t>
   Conformance criteria and considerations regarding error handling
   are defined in <xref target="HTTP" x:rel="#conformance"/>.
</t>
</section>

<section title="Syntax Notation" anchor="notation">
<iref primary="true" item="Grammar" subitem="ALPHA"/>
<iref primary="true" item="Grammar" subitem="CR"/>
<iref primary="true" item="Grammar" subitem="CRLF"/>
<iref primary="true" item="Grammar" subitem="CTL"/>
<iref primary="true" item="Grammar" subitem="DIGIT"/>
<iref primary="true" item="Grammar" subitem="DQUOTE"/>
<iref primary="true" item="Grammar" subitem="HEXDIG"/>
<iref primary="true" item="Grammar" subitem="HTAB"/>
<iref primary="true" item="Grammar" subitem="LF"/>
<iref primary="true" item="Grammar" subitem="OCTET"/>
<iref primary="true" item="Grammar" subitem="SP"/>
<iref primary="true" item="Grammar" subitem="VCHAR"/>
<t>
   This specification uses the Augmented Backus-Naur Form (ABNF) notation of
   <xref target="RFC5234"/>, extended with the notation for case-sensitivity
   in strings defined in <xref target="RFC7405"/>.
</t>
<t>
   It also uses a list extension, defined in <xref target="abnf.extension"/>,
   that allows for compact definition of comma-separated lists using a "#"
   operator (similar to how the "*" operator indicates repetition). <xref
   target="collected.abnf"/> shows the collected grammar with all list
   operators expanded to standard ABNF notation.
</t>
<t>
   As a convention, ABNF rule names prefixed with "obs-" denote
   obsolete grammar rules that appear for historical reasons.
</t>
<t anchor="core.rules">
  <x:anchor-alias value="ALPHA"/>
  <x:anchor-alias value="CR"/>
  <x:anchor-alias value="CRLF"/>
  <x:anchor-alias value="CTL"/>
  <x:anchor-alias value="DIGIT"/>
  <x:anchor-alias value="DQUOTE"/>
  <x:anchor-alias value="HEXDIG"/>
  <x:anchor-alias value="HTAB"/>
  <x:anchor-alias value="LF"/>
  <x:anchor-alias value="OCTET"/>
  <x:anchor-alias value="SP"/>
  <x:anchor-alias value="VCHAR"/>
   The following core rules are included by
   reference, as defined in <xref target="RFC5234" x:fmt="," x:sec="B.1"/>:
   ALPHA (letters), CR (carriage return), CRLF (CR LF), CTL (controls),
   DIGIT (decimal 0-9), DQUOTE (double quote),
   HEXDIG (hexadecimal 0-9/A-F/a-f), HTAB (horizontal tab), LF (line feed),
   OCTET (any 8-bit sequence of data), SP (space), and
   VCHAR (any visible <xref target="USASCII"/> character).
</t>
<t anchor="imported.rules">
  <x:anchor-alias value="absolute-path"/>
  <x:anchor-alias value="comment"/>
  <x:anchor-alias value="field-name"/>
  <x:anchor-alias value="field-value"/>
  <x:anchor-alias value="obs-text"/>
  <x:anchor-alias value="quoted-string"/>
  <x:anchor-alias value="token"/>
  The rules below are defined in <xref target="HTTP"/>:
</t>
<sourcecode type="abnf7230">
  <x:ref>BWS</x:ref>           = &lt;BWS, see <xref target="whitespace"/>&gt;
  <x:ref>OWS</x:ref>           = &lt;OWS, see <xref target="whitespace"/>&gt;
  <x:ref>RWS</x:ref>           = &lt;RWS, see <xref target="whitespace"/>&gt;
  <x:ref>absolute-path</x:ref> = &lt;absolute-path, see <xref target="uri.references"/>&gt;
  <x:ref>field-name</x:ref>    = &lt;field-name, see <xref target="fields.names"/>&gt;
  <x:ref>field-value</x:ref>   = &lt;field-value, see <xref target="fields.values"/>&gt;
  <x:ref>obs-text</x:ref>      = &lt;obs-text, see <xref target="quoted.strings"/>&gt;
  <x:ref>quoted-string</x:ref> = &lt;quoted-string, see <xref target="quoted.strings"/>&gt;
  <x:ref>token</x:ref>         = &lt;token, see <xref target="tokens"/>&gt;
  <x:ref>transfer-coding</x:ref> =
                  &lt;transfer-coding, see <xref target="field.te"/>&gt;
</sourcecode>
<t anchor="imported.uri.rules">
  <x:anchor-alias value="absolute-URI"/>
  <x:anchor-alias value="authority"/>
  <x:anchor-alias value="query"/>
  The rules below are defined in <xref target="URI"/>:
</t>
<sourcecode type="abnf7230">
  <x:ref>absolute-URI</x:ref>  = &lt;absolute-URI, see <xref target="URI" x:fmt="," x:sec="4.3"/>&gt;
  <x:ref>authority</x:ref>     = &lt;authority, see <xref target="URI" x:fmt="," x:sec="3.2"/>&gt;
  <x:ref>uri-host</x:ref>      = &lt;host, see <xref target="URI" x:fmt="," x:sec="3.2.2"/>&gt;
  <x:ref>port</x:ref>          = &lt;port, see <xref target="URI" x:fmt="," x:sec="3.2.3"/>&gt;
  <x:ref>query</x:ref>         = &lt;query, see <xref target="URI" x:fmt="," x:sec="3.4"/>&gt;
</sourcecode>
</section>
</section>

<section title="Message" anchor="http.message">
<t>
   HTTP/1.1 clients and servers communicate by sending messages.
   See <xref target="HTTP" x:fmt="of" x:rel="#terminology"/> for
   the general terminology and core concepts of HTTP.
</t>
<section title="Message Format" anchor="message.format">
<x:anchor-alias value="generic-message"/>
<x:anchor-alias value="message.types"/>
<x:anchor-alias value="HTTP-message"/>
<x:anchor-alias value="start-line"/>
<iref item="header section"/>
<iref item="headers"/>
<iref item="header line"/>
<t>
   An HTTP/1.1 message consists of a start-line followed by a CRLF and a
   sequence of
   octets in a format similar to the Internet Message Format
   <xref target="RFC5322"/>: zero or more header field lines (collectively
   referred to as the "headers" or the "header section"), an empty line
   indicating the end of the header section, and an optional message body.
</t>
<sourcecode type="abnf7230"><iref primary="true" item="Grammar" subitem="HTTP-message"><!--terminal production--></iref>
  <x:ref>HTTP-message</x:ref>   = <x:ref>start-line</x:ref> <x:ref>CRLF</x:ref>
                   *( <x:ref>field-line</x:ref> <x:ref>CRLF</x:ref> )
                   <x:ref>CRLF</x:ref>
                   [ <x:ref>message-body</x:ref> ]
</sourcecode>
<t>
   A message can be either a request from client to server or a
   response from server to client.  Syntactically, the two types of messages
   differ only in the start-line, which is either a request-line (for requests)
   or a status-line (for responses), and in the algorithm for determining
   the length of the message body (<xref target="message.body"/>).
</t>
<sourcecode type="abnf7230"><iref primary="true" item="Grammar" subitem="start-line"/>
  <x:ref>start-line</x:ref>     = <x:ref>request-line</x:ref> / <x:ref>status-line</x:ref>
</sourcecode>
<t>
   In theory, a client could receive requests and a server could receive
   responses, distinguishing them by their different start-line formats.
   In practice, servers are implemented to only expect a request
   (a response is interpreted as an unknown or invalid request method),
   and clients are implemented to only expect a response.
</t>
<t>
   HTTP makes use of some protocol elements similar to the
   Multipurpose Internet Mail Extensions (MIME) <xref target="RFC2045"/>.
   See <xref target="differences.between.http.and.mime"/> for the
   differences between HTTP and MIME messages.
</t>
</section>

<section title="Message Parsing" anchor="message.parsing">
<t>
   The normal procedure for parsing an HTTP message is to read the
   start-line into a structure, read each header field line into a hash
   table by field name until the empty line, and then use the parsed
   data to determine if a message body is expected.  If a message body
   has been indicated, then it is read as a stream until an amount
   of octets equal to the message body length is read or the connection
   is closed.
</t>
<t>
   A recipient &MUST; parse an HTTP message as a sequence of octets in an
   encoding that is a superset of US-ASCII <xref target="USASCII"/>.
   Parsing an HTTP message as a stream of Unicode characters, without regard
   for the specific encoding, creates security vulnerabilities due to the
   varying ways that string processing libraries handle invalid multibyte
   character sequences that contain the octet LF (%x0A).  String-based
   parsers can only be safely used within protocol elements after the element
   has been extracted from the message, such as within a header field line value
   after message parsing has delineated the individual field lines.
</t>
<t>
   Although the line terminator for the start-line and
   fields is the sequence CRLF, a recipient &MAY; recognize a
   single LF as a line terminator and ignore any preceding CR.
</t>
<t>
   A sender &MUST-NOT; generate a bare CR (a CR character not immediately
   followed by LF) within any protocol elements other than the content.
   A recipient of such a bare CR &MUST; consider that element to be invalid or
   replace each bare CR with SP before processing the element or forwarding
   the message.
</t>
<t>
   Older HTTP/1.0 user agent implementations might send an extra CRLF
   after a POST request as a workaround for some early server
   applications that failed to read message body content that was
   not terminated by a line-ending. An HTTP/1.1 user agent &MUST-NOT;
   preface or follow a request with an extra CRLF.  If terminating
   the request message body with a line-ending is desired, then the
   user agent &MUST; count the terminating CRLF octets as part of the
   message body length.
</t>
<t>
   In the interest of robustness, a server that is expecting to receive and
   parse a request-line &SHOULD; ignore at least one empty line (CRLF)
   received prior to the request-line.
</t>
<t>
   A sender &MUST-NOT; send whitespace between the start-line and
   the first header field.
</t>
<t>
   A recipient that receives whitespace between the start-line and
   the first header field &MUST; either reject the message as invalid or
   consume each whitespace-preceded line without further processing of it
   (i.e., ignore the entire line, along with any subsequent lines preceded
   by whitespace, until a properly formed header field is received or the
   header section is terminated).
   Rejection or removal of invalid whitespace-preceded lines is necessary
   to prevent their misinterpretation by downstream recipients that might
   be vulnerable to request smuggling (<xref target="request.smuggling"/>)
   or response splitting (<xref target="response.splitting"/>) attacks.
</t>
<t>
   When a server listening only for HTTP request messages, or processing
   what appears from the start-line to be an HTTP request message,
   receives a sequence of octets that does not match the HTTP-message
   grammar aside from the robustness exceptions listed above, the
   server &SHOULD; respond with a <x:ref>400 (Bad Request)</x:ref> response
   and close the connection.
</t>
</section>

<section title="HTTP Version" anchor="http.version">
  <x:anchor-alias value="HTTP-version"/>
  <x:anchor-alias value="HTTP-name"/>
<t>
   HTTP uses a "&lt;major&gt;.&lt;minor&gt;" numbering scheme to indicate
   versions of the protocol. This specification defines version "1.1".
   <xref target="protocol.version"/> specifies the semantics of HTTP version
   numbers.
</t>
<t>
   The version of an HTTP/1.x message is indicated by an HTTP-version field
   in the <x:ref>start-line</x:ref>. HTTP-version is case-sensitive.
</t>
<sourcecode type="abnf7230"><iref primary="true" item="Grammar" subitem="HTTP-version"/><iref primary="true" item="Grammar" subitem="HTTP-name"/>
  <x:ref>HTTP-version</x:ref>  = <x:ref>HTTP-name</x:ref> "/" <x:ref>DIGIT</x:ref> "." <x:ref>DIGIT</x:ref>
  <x:ref>HTTP-name</x:ref>     = %s"HTTP"
</sourcecode>
<t>
   When an HTTP/1.1 message is sent to an HTTP/1.0 recipient
   <xref target="HTTP10"/> or a recipient whose version is unknown,
   the HTTP/1.1 message is constructed such that it can be interpreted
   as a valid HTTP/1.0 message if all of the newer features are ignored.
   This specification places recipient-version requirements on some
   new features so that a conformant sender will only use compatible
   features until it has determined, through configuration or the
   receipt of a message, that the recipient supports HTTP/1.1.
</t>
<t>
   Intermediaries that process HTTP messages (i.e., all intermediaries
   other than those acting as tunnels) &MUST; send their own HTTP-version
   in forwarded messages, unless it is purposefully downgraded as a workaround
   for an upstream issue. In other words, an intermediary is not allowed to blindly
   forward the <x:ref>start-line</x:ref> without ensuring that the
   protocol version in that message matches a version to which that
   intermediary is conformant for both the receiving and
   sending of messages.  Forwarding an HTTP message without rewriting
   the HTTP-version might result in communication errors when downstream
   recipients use the message sender's version to determine what features
   are safe to use for later communication with that sender.
</t>
<t>
   A server &MAY; send an HTTP/1.0 response to an HTTP/1.1 request
   if it is known or suspected that the client incorrectly implements the
   HTTP specification and is incapable of correctly processing later
   version responses, such as when a client fails to parse the version
   number correctly or when an intermediary is known to blindly forward
   the HTTP-version even when it doesn't conform to the given minor
   version of the protocol. Such protocol downgrades &SHOULD-NOT; be
   performed unless triggered by specific client attributes, such as when
   one or more of the request header fields (e.g., <x:ref>User-Agent</x:ref>)
   uniquely match the values sent by a client known to be in error.
</t>
</section>
</section>

<section title="Request Line" anchor="request.line">
  <x:anchor-alias value="Request"/>
  <x:anchor-alias value="request-line"/>
<t>
   A request-line begins with a method token, followed by a single
   space (SP), the request-target, and another single space (SP), and ends
   with the protocol version.
</t>
<sourcecode type="abnf7230"><iref primary="true" item="Grammar" subitem="request-line"/>
  <x:ref>request-line</x:ref>   = <x:ref>method</x:ref> <x:ref>SP</x:ref> <x:ref>request-target</x:ref> <x:ref>SP</x:ref> <x:ref>HTTP-version</x:ref>
</sourcecode>
<t>
   Although the request-line grammar rule requires that each of the component
   elements be separated by a single SP octet, recipients &MAY; instead parse
   on whitespace-delimited word boundaries and, aside from the CRLF
   terminator, treat any form of whitespace as the SP separator while
   ignoring preceding or trailing whitespace; such whitespace includes one or
   more of the following octets: SP, HTAB, VT (%x0B), FF (%x0C), or bare CR.
   However, lenient parsing can result in request smuggling security
   vulnerabilities if there are multiple recipients of the message and each
   has its own unique interpretation of robustness
   (see <xref target="request.smuggling"/>).
</t>
<t>
   HTTP does not place a predefined limit on the length of a request-line,
   as described in <xref target="HTTP" x:rel="#length.requirements"/>.
   A server that receives a method longer than any that it implements
   &SHOULD; respond with a <x:ref>501 (Not Implemented)</x:ref> status code.
   A server that receives a request-target longer than any URI it wishes to
   parse &MUST; respond with a
   <x:ref>414 (URI Too Long)</x:ref> status code (see <xref target="status.414"/>).
</t>
<t>
   Various ad hoc limitations on request-line length are found in practice.
   It is &RECOMMENDED; that all HTTP senders and recipients support, at a
   minimum, request-line lengths of 8000 octets.
</t>

<section title="Method" anchor="request.method">
  <iref primary="true" item="method"/>
  <x:anchor-alias value="method"/>
<t>
   The method token indicates the request method to be performed on the
   target resource. The request method is case-sensitive.
</t>
<sourcecode type="abnf7230"><iref primary="true" item="Grammar" subitem="method"/>
  <x:ref>method</x:ref>         = <x:ref>token</x:ref>
</sourcecode>
<t>
   The request methods defined by this specification can be found in
   <xref target="methods"/>, along with information regarding the HTTP method
   registry and considerations for defining new methods.
</t>
</section>

<section title="Request Target" anchor="request.target">
  <iref primary="true" item="request-target"/>
  <x:anchor-alias value="request-target"/>
<t>
   The request-target identifies the target resource upon which to apply the
   request. The client derives a request-target from its desired target URI.
   There are four distinct formats for the request-target, depending on both
   the method being requested and whether the request is to a proxy.
</t>
<sourcecode type="abnf7230"><iref primary="true" item="Grammar" subitem="request-target"/><iref primary="false" item="Grammar" subitem="origin-form"/><iref primary="false" item="Grammar" subitem="absolute-form"/><iref primary="false" item="Grammar" subitem="authority-form"/><iref primary="false" item="Grammar" subitem="asterisk-form"/>
  <x:ref>request-target</x:ref> = <x:ref>origin-form</x:ref>
                 / <x:ref>absolute-form</x:ref>
                 / <x:ref>authority-form</x:ref>
                 / <x:ref>asterisk-form</x:ref>
</sourcecode>
<t>
   No whitespace is allowed in the request-target.
   Unfortunately, some user agents fail to properly encode or exclude
   whitespace found in hypertext references, resulting in those disallowed
   characters being sent as the request-target in a malformed request-line.
</t>
<t>
   Recipients of an invalid request-line &SHOULD; respond with either a
   <x:ref>400 (Bad Request)</x:ref> error or a <x:ref>301 (Moved Permanently)</x:ref>
   redirect with the request-target properly encoded.  A recipient &SHOULD-NOT;
   attempt to autocorrect and then process the request without a redirect,
   since the invalid request-line might be deliberately crafted to bypass
   security filters along the request chain.
</t>
<t>
   A client &MUST; send a <x:ref>Host</x:ref> header field (<xref target="field.host"/>)
   in all HTTP/1.1 request messages.
   If the target URI includes an authority component, then a client &MUST;
   send a field value for Host that is identical to that authority
   component, excluding any userinfo subcomponent and its "@" delimiter
   (<xref target="uri.schemes"/>).
   If the authority component is missing or undefined for the target URI,
   then a client &MUST; send a Host header field with an empty field value.
</t>
<t>
   A server &MUST; respond with a <x:ref>400 (Bad Request)</x:ref> status code
   to any HTTP/1.1 request message that lacks a Host header field and
   to any request message that contains more than one Host header field line
   or a Host header field with an invalid field value.
</t>

<section title="origin-form" anchor="origin-form">
   <iref item="origin-form (of request-target)"/>
<t>
   The most common form of request-target is the <x:dfn>origin-form</x:dfn>.
</t>
<sourcecode type="abnf7230"><iref primary="true" item="Grammar" subitem="origin-form"/>
  <x:ref>origin-form</x:ref>    = <x:ref>absolute-path</x:ref> [ "?" <x:ref>query</x:ref> ]
</sourcecode>
<t>
   When making a request directly to an origin server, other than a CONNECT
   or server-wide OPTIONS request (as detailed below),
   a client &MUST; send only the absolute path and query components of
   the target URI as the request-target.
   If the target URI's path component is empty, the client &MUST; send
   "/" as the path within the origin-form of request-target.
   A <x:ref>Host</x:ref> header field is also sent, as defined in
   <xref target="field.host"/>.
</t>
<t>
   For example, a client wishing to retrieve a representation of the resource
   identified as
</t>
<artwork x:indent-with="  " type="example">
http://www.example.org/where?q=now
</artwork>
<t>
   directly from the origin server would open (or reuse) a TCP connection
   to port 80 of the host "www.example.org" and send the lines:
</t>
<sourcecode type="http-message">
GET /where?q=now HTTP/1.1
Host: www.example.org
</sourcecode>
<t>
   followed by the remainder of the request message.
</t>
</section>

<section title="absolute-form" anchor="absolute-form">
   <iref item="absolute-form (of request-target)"/>
<t>
   When making a request to a proxy, other than a CONNECT or server-wide
   OPTIONS request (as detailed below), a client &MUST; send the target URI
   in <x:dfn>absolute-form</x:dfn> as the request-target.
</t>
<sourcecode type="abnf7230"><iref primary="true" item="Grammar" subitem="absolute-form"/>
  <x:ref>absolute-form</x:ref>  = <x:ref>absolute-URI</x:ref>
</sourcecode>
<t>
   The proxy is requested to either service that request from a valid cache,
   if possible, or make the same request on the client's behalf either to
   the next inbound proxy server or directly to the origin server indicated
   by the request-target.  Requirements on such "forwarding" of messages are
   defined in <xref target="message.forwarding"/>.
</t>
<t>
   An example absolute-form of request-line would be:
</t>
<sourcecode type="http-message">
GET http://www.example.org/pub/WWW/TheProject.html HTTP/1.1
</sourcecode>
<t>
   A client &MUST; send a Host header field in an HTTP/1.1 request even
   if the request-target is in the absolute-form, since this
   allows the Host information to be forwarded through ancient HTTP/1.0
   proxies that might not have implemented Host.
</t>
<t>
   When a proxy receives a request with an absolute-form of
   request-target, the proxy &MUST; ignore the received
   Host header field (if any) and instead replace it with the host
   information of the request-target.  A proxy that forwards such a request
   &MUST; generate a new Host field value based on the received
   request-target rather than forward the received Host field value.
</t>
<t>
   When an origin server receives a request with an absolute-form of
   request-target, the origin server &MUST; ignore the received Host header
   field (if any) and instead use the host information of the request-target.
   Note that if the request-target does not have an authority component, an
   empty Host header field will be sent in this case.
</t>
<t>
   A server &MUST; accept the absolute-form in requests even though most
   HTTP/1.1 clients will only send the absolute-form to a proxy.
</t>
</section>

<section title="authority-form" anchor="authority-form">
   <iref item="authority-form (of request-target)"/>
<t>
   The <x:dfn>authority-form</x:dfn> of request-target is only used for
   CONNECT requests (<xref target="CONNECT"/>). It consists of only the
   <x:ref>uri-host</x:ref> and <x:ref>port</x:ref> number of the tunnel
   destination, separated by a colon (":").
</t>
<sourcecode type="abnf7230"><iref primary="true" item="Grammar" subitem="authority-form"/>
  <x:ref>authority-form</x:ref> = <x:ref>uri-host</x:ref> ":" <x:ref>port</x:ref>
</sourcecode>
<t>
   When making a CONNECT request to establish a tunnel through one or more
   proxies, a client &MUST; send only the host and port of the tunnel
   destination as the request-target. The client obtains the host and port
   from the target URI's <x:ref>authority</x:ref> component, except that it
   sends the scheme's default port if the target URI elides the port.
   For example, a CONNECT request to "http://www.example.com" looks like the
   following:
</t>
<sourcecode type="http-message">
CONNECT www.example.com:80 HTTP/1.1
Host: www.example.com

</sourcecode>
</section>

<section title="asterisk-form" anchor="asterisk-form">
   <iref item="asterisk-form (of request-target)"/>
<t>
   The <x:dfn>asterisk-form</x:dfn> of request-target is only used for a server-wide
   OPTIONS request (<xref target="OPTIONS"/>).
</t>
<sourcecode type="abnf7230"><iref primary="true" item="Grammar" subitem="asterisk-form"/>
  <x:ref>asterisk-form</x:ref>  = "*"
</sourcecode>
<t>
   When a client wishes to request OPTIONS
   for the server as a whole, as opposed to a specific named resource of
   that server, the client &MUST; send only "*" (%x2A) as the request-target.
   For example,
</t>
<sourcecode type="http-message">
OPTIONS * HTTP/1.1
</sourcecode>
<t>
   If a proxy receives an OPTIONS request with an absolute-form of
   request-target in which the URI has an empty path and no query component,
   then the last proxy on the request chain &MUST; send a request-target
   of "*" when it forwards the request to the indicated origin server.
</t>
<t>
   For example, the request
</t>
<sourcecode type="http-message">
OPTIONS http://www.example.org:8001 HTTP/1.1
</sourcecode>
<t>
  would be forwarded by the final proxy as
</t>
<sourcecode type="http-message">
OPTIONS * HTTP/1.1
Host: www.example.org:8001
</sourcecode>
<t>
   after connecting to port 8001 of host "www.example.org".
</t>
</section>
</section>

<section title="Reconstructing the Target URI" anchor="reconstructing.target.uri">
  <x:anchor-alias value="h1.effective.request.uri"/>
<t>
   The target URI is the <x:ref>request-target</x:ref> when the
   request-target is in <x:ref>absolute-form</x:ref>. In that case,
   a server will parse the URI into its generic components for further
   evaluation.
</t>
<t>
   Otherwise, the server reconstructs the target URI from the connection
   context and various parts of the request message in order to identify the
   target resource (<xref target="target.resource"/>):
</t>
<ul>
<li>
   If the server's configuration provides for a fixed URI scheme, or a scheme
   is provided by a trusted outbound gateway, that scheme is used for the
   target URI. This is common in large-scale deployments because a gateway
   server will receive the client's connection context and replace that with
   their own connection to the inbound server.
   Otherwise, if the request is received over a secured connection,
   the target URI's scheme is "https"; if not, the scheme is "http".
</li>
<li>
   If the request-target is in <x:ref>authority-form</x:ref>,
   the target URI's authority component is the request-target.
   Otherwise, the target URI's authority component is the field value of the
   <x:ref>Host</x:ref> header field. If there is no <x:ref>Host</x:ref>
   header field or if its field value is empty or invalid,
   the target URI's authority component is empty.
</li>
<li>
   If the request-target is in <x:ref>authority-form</x:ref> or
   <x:ref>asterisk-form</x:ref>, the target URI's combined
   <x:ref>path</x:ref> and <x:ref>query</x:ref> component is empty.
   Otherwise, the target URI's combined <x:ref>path</x:ref> and
   <x:ref>query</x:ref> component is the request-target.
</li>
<li>
   The components of a reconstructed target URI, once determined as above,
   can be recombined into <x:ref>absolute-URI</x:ref> form by concatenating
   the scheme, "://", authority, and combined path and query component.
</li>
</ul>
<t>
   Example 1: The following message received over a secure connection
</t>
<sourcecode type="http-message">
GET /pub/WWW/TheProject.html HTTP/1.1
Host: www.example.org
</sourcecode>
<t>
  has a target URI of
</t>
<artwork type="example" x:indent-with="  ">
https://www.example.org/pub/WWW/TheProject.html
</artwork>
<t>
   Example 2: The following message received over an insecure connection
</t>
<sourcecode type="http-message">
OPTIONS * HTTP/1.1
Host: www.example.org:8080
</sourcecode>
<t>
  has a target URI of
</t>
<artwork type="example" x:indent-with="  ">
http://www.example.org:8080
</artwork>
<t>
   If the target URI's authority component is empty and its URI scheme
   requires a non-empty authority (as is the case for "http" and "https"),
   the server can reject the request or determine whether a configured
   default applies that is consistent with the incoming connection's context.
   Context might include connection details like address and port, what
   security has been applied, and locally defined information specific to
   that server's configuration. An empty authority is replaced with the
   configured default before further processing of the request.
</t>
<t>
   Supplying a default name for authority within the context of a secured
   connection is inherently unsafe if there is any chance that the user
   agent's intended authority might differ from the default.
   A server that can uniquely identify an authority from the request
   context &MAY; use that identity as a default without this risk.
   Alternatively, it might be better to redirect the request to a safe
   resource that explains how to obtain a new client.
</t>
<t>
   Note that reconstructing the client's target URI is only half of the
   process for identifying a target resource. The other half is determining
   whether that target URI identifies a resource for which the server is
   willing and able to send a response, as defined in
   <xref target="routing.reject"/>.
</t>
</section>
</section>

<section title="Status Line" anchor="status.line">
  <x:anchor-alias value="response"/>
  <x:anchor-alias value="status-line"/>
  <x:anchor-alias value="status-code"/>
  <x:anchor-alias value="status code"/>
  <x:anchor-alias value="reason-phrase"/>
<t>
   The first line of a response message is the status-line, consisting
   of the protocol version, a space (SP), the status code, and another space
   and ending with an &OPTIONAL; textual phrase describing the status code.
</t>
<sourcecode type="abnf7230"><iref primary="true" item="Grammar" subitem="status-line"/>
  <x:ref>status-line</x:ref> = <x:ref>HTTP-version</x:ref> <x:ref>SP</x:ref> <x:ref>status-code</x:ref> <x:ref>SP</x:ref> [ <x:ref>reason-phrase</x:ref> ]
</sourcecode>
<t>
   Although the status-line grammar rule requires that each of the component
   elements be separated by a single SP octet, recipients &MAY; instead parse
   on whitespace-delimited word boundaries and, aside from the line
   terminator, treat any form of whitespace as the SP separator while
   ignoring preceding or trailing whitespace; such whitespace includes one or
   more of the following octets: SP, HTAB, VT (%x0B), FF (%x0C), or bare CR.
   However, lenient parsing can result in response splitting security
   vulnerabilities if there are multiple recipients of the message and each
   has its own unique interpretation of robustness
   (see <xref target="response.splitting"/>).
</t>
<t>
   The status-code element is a 3-digit integer code describing the
   result of the server's attempt to understand and satisfy the client's
   corresponding request. A recipient parses and interprets the remainder
   of the response message in light of the semantics defined for that
   status code, if the status code is recognized by that recipient,
   or in accordance with the class of that status code when the specific
   code is unrecognized.
</t>
<sourcecode type="abnf7230"><iref primary="true" item="Grammar" subitem="status-code"/>
  <x:ref>status-code</x:ref>    = 3<x:ref>DIGIT</x:ref>
</sourcecode>
<t>
   HTTP's core status codes are defined in <xref target="status.codes"/>,
   along with the classes of status codes, considerations for the
   definition of new status codes, and the IANA registry for collecting
   such definitions.
</t>
<t>
   The reason-phrase element exists for the sole purpose of providing a
   textual description associated with the numeric status code, mostly out of
   deference to earlier Internet application protocols that were more
   frequently used with interactive text clients.
</t>
<sourcecode type="abnf7230"><iref primary="true" item="Grammar" subitem="reason-phrase"/>
  <x:ref>reason-phrase</x:ref>  = 1*( <x:ref>HTAB</x:ref> / <x:ref>SP</x:ref> / <x:ref>VCHAR</x:ref> / <x:ref>obs-text</x:ref> )
</sourcecode>
<t>
   A client &SHOULD; ignore the reason-phrase content because it is not a
   reliable channel for information (it might be translated for a given locale,
   overwritten by intermediaries, or discarded when the message is forwarded
   via other versions of HTTP).
   A server &MUST; send the space that separates the status-code from the
   reason-phrase even when the reason-phrase is absent (i.e., the status-line
   would end with the space).
</t>
</section>

<section title="Field Syntax" anchor="header.field.syntax">
  <x:anchor-alias value="field-line"/>
<t>
   Each field line consists of a case-insensitive field name
   followed by a colon (":"), optional leading whitespace, the field line value,
   and optional trailing whitespace.
</t>
<sourcecode type="abnf7230"><iref primary="true" item="Grammar" subitem="field-line"/><iref primary="false" item="Grammar" subitem="field-name"/><iref primary="false" item="Grammar" subitem="field-value"/>
  <x:ref>field-line</x:ref>   = <x:ref>field-name</x:ref> ":" <x:ref>OWS</x:ref> <x:ref>field-value</x:ref> <x:ref>OWS</x:ref>
</sourcecode>
<t>
   Rules for parsing within field values are
   defined in <xref target="fields.values"/>. This section covers the
   generic syntax for header field inclusion within, and extraction from,
   HTTP/1.1 messages.
</t>

<section title="Field Line Parsing" anchor="field.parsing">
  <x:anchor-alias value="header.parsing"/>
<t>
   Messages are parsed using a generic algorithm, independent of the
   individual field names. The contents within a given field line value are
   not parsed until a later stage of message interpretation (usually after the
   message's entire field section has been processed).
</t>
<t>
   No whitespace is allowed between the field name and colon.
   In the past, differences in the handling of such whitespace have led to
   security vulnerabilities in request routing and response handling.
   A server &MUST; reject, with a response status code of
   <x:ref>400 (Bad Request)</x:ref>, any received request message that contains
   whitespace between a header field name and colon.
   A proxy &MUST; remove any such whitespace
   from a response message before forwarding the message downstream.
</t>
<t>
   A field line value might be preceded and/or followed by optional whitespace
   (OWS); a single SP preceding the field line value is preferred for consistent
   readability by humans.
   The field line value does not include that leading or trailing whitespace: OWS
   occurring before the first non-whitespace octet of the field line value,
   or after the last non-whitespace octet of the field line value, is excluded by
   parsers when extracting the field line value from a field line.
</t>
</section>

<section title="Obsolete Line Folding" anchor="line.folding">
  <x:anchor-alias value="obs-fold"/>
<t>
   Historically, HTTP/1.x field values could be extended over multiple
   lines by preceding each extra line with at least one space or horizontal
   tab (obs-fold). This specification deprecates such line folding except
   within the "message/http" media type
   (<xref target="media.type.message.http"/>).
</t>
<sourcecode type="abnf7230"><iref primary="true" item="Grammar" subitem="obs-fold"/>
  <x:ref>obs-fold</x:ref>     = <x:ref>OWS</x:ref> <x:ref>CRLF</x:ref> <x:ref>RWS</x:ref>
               ; obsolete line folding
</sourcecode>
<t>
   A sender &MUST-NOT; generate a message that includes line folding
   (i.e., that has any field line value that contains a match to the
   <x:ref>obs-fold</x:ref> rule) unless the message is intended for packaging
   within the "message/http" media type.
</t>
<t>
   A server that receives an <x:ref>obs-fold</x:ref> in a request message that
   is not within a "message/http" container &MUST; either reject the message by
   sending a <x:ref>400 (Bad Request)</x:ref>, preferably with a
   representation explaining that obsolete line folding is unacceptable, or
   replace each received <x:ref>obs-fold</x:ref> with one or more
   <x:ref>SP</x:ref> octets prior to interpreting the field value or
   forwarding the message downstream.
</t>
<t>
   A proxy or gateway that receives an <x:ref>obs-fold</x:ref> in a response
   message that is not within a "message/http" container &MUST; either discard
   the message and replace it with a <x:ref>502 (Bad Gateway)</x:ref>
   response, preferably with a representation explaining that unacceptable
   line folding was received, or replace each received <x:ref>obs-fold</x:ref>
   with one or more <x:ref>SP</x:ref> octets prior to interpreting the field
   value or forwarding the message downstream.
</t>
<t>
   A user agent that receives an <x:ref>obs-fold</x:ref> in a response message
   that is not within a "message/http" container &MUST; replace each received
   <x:ref>obs-fold</x:ref> with one or more <x:ref>SP</x:ref> octets prior to
   interpreting the field value.
</t>
</section>
</section>

<section title="Message Body" anchor="message.body">
  <x:anchor-alias value="message-body"/>
<t>
   The message body (if any) of an HTTP/1.1 message is used to carry content
   (<xref target="content"/>) for the request or response. The
   message body is identical to the content unless a transfer coding has
   been applied, as described in <xref target="field.transfer-encoding"/>.
</t>
<sourcecode type="abnf7230"><iref primary="true" item="Grammar" subitem="message-body"/>
  <x:ref>message-body</x:ref> = *OCTET
</sourcecode>
<t>
   The rules for determining when a message body is present in an HTTP/1.1
   message differ for requests and responses.
</t>
<t>
   The presence of a message body in a request is signaled by a
   <x:ref>Content-Length</x:ref> or <x:ref>Transfer-Encoding</x:ref> header
   field. Request message framing is independent of method semantics.
</t>
<t>
   The presence of a message body in a response, as detailed in
   <xref target="message.body.length"/>, depends on both the request
   method to which it is responding and the response status code. This
   corresponds to when response content is allowed by HTTP semantics
   (<xref target="content.semantics"/>).
</t>

<section title="Transfer-Encoding" anchor="field.transfer-encoding">
  <x:anchor-alias value="header.transfer-encoding"/>
  <iref primary="true" item="Fields" subitem="Transfer-Encoding" x:for-anchor=""/><iref primary="true" item="Header Fields" subitem="Transfer-Encoding" x:for-anchor=""/><iref primary="true" item="Transfer-Encoding header field" x:for-anchor=""/>
  <iref item="chunked (Coding Format)"/>
  <x:anchor-alias value="Transfer-Encoding"/>
<t>
   The Transfer-Encoding header field lists the transfer coding names
   corresponding to the sequence of transfer codings that have been
   (or will be) applied to the content in order to form the message body.
   Transfer codings are defined in <xref target="transfer.codings"/>.
</t>
<sourcecode type="abnf7230"><iref primary="true" item="Grammar" subitem="Transfer-Encoding"/>
  <x:ref>Transfer-Encoding</x:ref> = #<x:ref>transfer-coding</x:ref>
                       ; defined in <xref target="field.te"/>
</sourcecode>
<t>
   Transfer-Encoding is analogous to the Content-Transfer-Encoding field of
   MIME, which was designed to enable safe transport of binary data over a
   7-bit transport service (<xref target="RFC2045" x:fmt="," x:sec="6"/>).
   However, safe transport has a different focus for an 8bit-clean transfer
   protocol. In HTTP's case, Transfer-Encoding is primarily intended to
   accurately delimit dynamically generated content. It also serves to
   distinguish encodings that are only applied in transit from the encodings
   that are a characteristic of the selected representation.
</t>
<t>
   A recipient &MUST; be able to parse the chunked transfer coding
   (<xref target="chunked.encoding"/>) because it plays a crucial role in
   framing messages when the content size is not known in advance.
   A sender &MUST-NOT; apply the chunked transfer coding more than once to a
   message body (i.e., chunking an already chunked message is not allowed).
   If any transfer coding other than chunked is applied to a request's content,
   the sender &MUST; apply chunked as the final transfer coding to
   ensure that the message is properly framed.
   If any transfer coding other than chunked is applied to a response's content,
   the sender &MUST; either apply chunked as the final transfer coding
   or terminate the message by closing the connection.
</t>
<t>
   For example,
</t>
<sourcecode type="http-message">
Transfer-Encoding: gzip, chunked
</sourcecode>
<t>
   indicates that the content has been compressed using the gzip
   coding and then chunked using the chunked coding while forming the
   message body.
</t>
<t>
   Unlike <x:ref>Content-Encoding</x:ref> (<xref target="content.codings"/>),
   Transfer-Encoding is a property of the message, not of the representation.
   Any recipient along the request/response chain &MAY; decode the received
   transfer coding(s) or apply additional transfer coding(s) to the message
   body, assuming that corresponding changes are made to the Transfer-Encoding
   field value. Additional information about the encoding parameters can be
   provided by other header fields not defined by this specification.
</t>
<t>
   Transfer-Encoding &MAY; be sent in a response to a HEAD request or in a
   <x:ref>304 (Not Modified)</x:ref> response (<xref target="status.304"/>) to a GET request,
   neither of which includes a message body,
   to indicate that the origin server would have applied a transfer coding
   to the message body if the request had been an unconditional GET.
   This indication is not required, however, because any recipient on
   the response chain (including the origin server) can remove transfer
   codings when they are not needed.
</t>
<t>
   A server &MUST-NOT; send a Transfer-Encoding header field in any response
   with a status code of
   <x:ref>1xx (Informational)</x:ref> or <x:ref>204 (No Content)</x:ref>.
   A server &MUST-NOT; send a Transfer-Encoding header field in any
   <x:ref>2xx (Successful)</x:ref> response to a CONNECT request (<xref target="CONNECT"/>).
</t>
<t>
   A server that receives a request message with a transfer coding it does
   not understand &SHOULD; respond with <x:ref>501 (Not Implemented)</x:ref>.
</t>
<t>
   Transfer-Encoding was added in HTTP/1.1. It is generally assumed that
   implementations advertising only HTTP/1.0 support will not understand
   how to process transfer-encoded content, and that an HTTP/1.0 message
   received with a Transfer-Encoding is likely to have been forwarded
   without proper handling of the chunked transfer coding in transit.
</t>
<t>
   A client &MUST-NOT; send a request containing Transfer-Encoding unless it
   knows the server will handle HTTP/1.1 requests (or later minor revisions);
   such knowledge might be in the form of specific user configuration or by
   remembering the version of a prior received response.
   A server &MUST-NOT; send a response containing Transfer-Encoding unless
   the corresponding request indicates HTTP/1.1 (or later minor revisions).
</t>
<t>
   Early implementations of Transfer-Encoding would occasionally send both
   a chunked transfer coding for message framing and an estimated Content-Length
   header field for use by progress bars. This is why Transfer-Encoding is
   defined as overriding Content-Length, as opposed to them being mutually
   incompatible. Unfortunately, forwarding such a message can lead to
   vulnerabilities regarding
   request smuggling (<xref target="request.smuggling"/>) or
   response splitting (<xref target="response.splitting"/>) attacks
   if any downstream recipient fails to parse the message according to this
   specification, particularly when a downstream recipient only implements
   HTTP/1.0.
</t>
<t>
   A server &MAY; reject a request that contains both Content-Length and
   Transfer-Encoding or process such a request in accordance with the
   Transfer-Encoding alone. Regardless, the server &MUST; close the connection
   after responding to such a request to avoid the potential attacks.
</t>
<t>
   A server or client that receives an HTTP/1.0 message containing a
   Transfer-Encoding header field &MUST; treat the message as if the framing
   is faulty, even if a Content-Length is present, and close the connection
   after processing the message. The message sender might have retained a
   portion of the message, in buffer, that could be misinterpreted by further
   use of the connection.
</t>
</section>

<section title="Content-Length" anchor="body.content-length">
  <iref primary="false" item="Content-Length header field" x:for-anchor=""/>
  <x:anchor-alias value="Content-Length"/>
<t>
   When a message does not have a <x:ref>Transfer-Encoding</x:ref> header
   field, a Content-Length header field (<xref target="field.content-length"/>) can provide the anticipated size,
   as a decimal number of octets, for potential content.
   For messages that do include content, the Content-Length field value
   provides the framing information necessary for determining where the data
   (and message) ends.  For messages that do not include content, the
   Content-Length indicates the size of the selected representation
   (<xref target="field.content-length"/>).
</t>
<t>
   A sender &MUST-NOT; send a Content-Length header field in any message that
   contains a <x:ref>Transfer-Encoding</x:ref> header field.
</t>
<aside>
  <t>
   &Note; HTTP's use of Content-Length for message framing differs
   significantly from the same field's use in MIME, where it is an optional
   field used only within the "message/external-body" media-type.
  </t>
</aside>
</section>

<section title="Message Body Length" anchor="message.body.length">
  <iref item="chunked (Coding Format)"/>
<t>
   The length of a message body is determined by one of the following
   (in order of precedence):
</t>
<ol>
  <li><t>
   Any response to a HEAD request and any response with a
   <x:ref>1xx (Informational)</x:ref>, <x:ref>204 (No Content)</x:ref>, or
   <x:ref>304 (Not Modified)</x:ref> status code is always
   terminated by the first empty line after the header fields, regardless of
   the header fields present in the message, and thus cannot contain a
   message body or trailer section.
  </t></li>
  <li><t>
   Any <x:ref>2xx (Successful)</x:ref> response to a CONNECT request implies that the
   connection will become a tunnel immediately after the empty line that
   concludes the header fields.  A client &MUST; ignore any
   <x:ref>Content-Length</x:ref> or <x:ref>Transfer-Encoding</x:ref> header
   fields received in such a message.
  </t></li>
  <li><t>
   If a message is received with both a <x:ref>Transfer-Encoding</x:ref>
   and a <x:ref>Content-Length</x:ref> header field, the Transfer-Encoding
   overrides the Content-Length. Such a message might indicate an attempt to
   perform request smuggling (<xref target="request.smuggling"/>) or
   response splitting (<xref target="response.splitting"/>) and ought to be
   handled as an error.
   An intermediary that chooses to forward the message &MUST; first remove the
   received Content-Length field and process the Transfer-Encoding
   (as described below) prior to forwarding the message downstream.
  </t></li>
  <li><t>
   If a <x:ref>Transfer-Encoding</x:ref> header field is present
   and the chunked transfer coding (<xref target="chunked.encoding"/>)
   is the final encoding, the message body length is determined by reading
   and decoding the chunked data until the transfer coding indicates the
   data is complete.
  </t>
  <t>
   If a <x:ref>Transfer-Encoding</x:ref> header field is present in a
   response and the chunked transfer coding is not the final encoding, the
   message body length is determined by reading the connection until it is
   closed by the server.
  </t>
  <t>
   If a <x:ref>Transfer-Encoding</x:ref> header field is present in a request
   and the chunked transfer coding is not the final encoding, the message body
   length cannot be determined reliably; the server &MUST; respond with
   the <x:ref>400 (Bad Request)</x:ref> status code and then close the
   connection.
  </t></li>
  <li><t>
   If a message is received without <x:ref>Transfer-Encoding</x:ref> and with
   an invalid <x:ref>Content-Length</x:ref> header field, then the message
   framing is invalid and the recipient &MUST; treat it as an unrecoverable
   error, unless the field value can be successfully parsed as a
   comma-separated list (<xref target="abnf.extension"/>), all values in the
   list are valid, and all values in the list are the same (in which case, the
   message is processed with that single value used as the Content-Length field
   value).
   If the unrecoverable error is in a request message,
   the server &MUST; respond with
   a <x:ref>400 (Bad Request)</x:ref> status code and then close the connection.
   If it is in a response message received by a proxy,
   the proxy &MUST; close the connection to the server, discard the received
   response, and send a <x:ref>502 (Bad Gateway)</x:ref> response to the
   client.
   If it is in a response message received by a user agent,
   the user agent &MUST; close the connection to the server and discard the
   received response.
  </t></li>
  <li><t>
   If a valid <x:ref>Content-Length</x:ref> header field is present without
   <x:ref>Transfer-Encoding</x:ref>, its decimal value defines the
   expected message body length in octets.
   If the sender closes the connection or the recipient times out before the
   indicated number of octets are received, the recipient &MUST; consider
   the message to be incomplete and close the connection.
  </t></li>
  <li><t>
   If this is a request message and none of the above are true, then the
   message body length is zero (no message body is present).
  </t></li>
  <li><t>
   Otherwise, this is a response message without a declared message body
   length, so the message body length is determined by the number of octets
   received prior to the server closing the connection.
  </t></li>
</ol>
<t>
   Since there is no way to distinguish a successfully completed,
   close-delimited response message from a partially received message interrupted
   by network failure, a server &SHOULD; generate encoding or
   length-delimited messages whenever possible.  The close-delimiting
   feature exists primarily for backwards compatibility with HTTP/1.0.
</t>
<aside>
  <t>
     &Note; Request messages are never close-delimited because they are always
     explicitly framed by length or transfer coding, with the absence of both implying
     the request ends immediately after the header section.
  </t>
</aside>
<t>
   A server &MAY; reject a request that contains a message body but
   not a <x:ref>Content-Length</x:ref> by responding with
   <x:ref>411 (Length Required)</x:ref>.
</t>
<t>
   Unless a transfer coding other than chunked has been applied,
   a client that sends a request containing a message body &SHOULD;
   use a valid <x:ref>Content-Length</x:ref> header field if the message body
   length is known in advance, rather than the chunked transfer coding, since some
   existing services respond to chunked with a <x:ref>411 (Length Required)</x:ref>
   status code even though they understand the chunked transfer coding.  This
   is typically because such services are implemented via a gateway that
   requires a content length in advance of being called, and the server
   is unable or unwilling to buffer the entire request before processing.
</t>
<t>
   A user agent that sends a request that contains a message body &MUST; send
   either a valid <x:ref>Content-Length</x:ref> header field or use the
   chunked transfer coding. A client &MUST-NOT; use the chunked transfer
   coding unless it knows the server will handle HTTP/1.1 (or later)
   requests; such knowledge can be in the form of specific user configuration
   or by remembering the version of a prior received response.
</t>
<t>
   If the final response to the last request on a connection has been
   completely received and there remains additional data to read, a user agent
   &MAY; discard the remaining data or attempt to determine if that data
   belongs as part of the prior message body, which might be the case if the
   prior message's Content-Length value is incorrect. A client &MUST-NOT;
   process, cache, or forward such extra data as a separate response, since
   such behavior would be vulnerable to cache poisoning.
</t>
</section>
</section>

<section title="Transfer Codings" anchor="transfer.codings">
<t>
   Transfer coding names are used to indicate an encoding
   transformation that has been, can be, or might need to be applied to a
   message's content in order to ensure "safe transport" through the network.
   This differs from a content coding in that the transfer coding is a
   property of the message rather than a property of the representation
   that is being transferred.
</t>
<t>
   All transfer-coding names are case-insensitive and ought to be registered
   within the HTTP Transfer Coding registry, as defined in
   <xref target="transfer.coding.registry"/>.
   They are used in the <x:ref>Transfer-Encoding</x:ref>
   (<xref target="field.transfer-encoding"/>) and <x:ref>TE</x:ref>
   (<xref target="field.te"/>) header fields (the latter also
   defining the "transfer-coding" grammar).
</t>

<section title="Chunked Transfer Coding" anchor="chunked.encoding">
  <iref primary="true" item="chunked (transfer coding)"/>
  <x:anchor-alias value="chunk"/>
  <x:anchor-alias value="chunked-body"/>
  <x:anchor-alias value="chunk-data"/>
  <x:anchor-alias value="chunk-size"/>
  <x:anchor-alias value="last-chunk"/>
<t>
   The chunked transfer coding wraps content in order to transfer it
   as a series of chunks, each with its own size indicator, followed by an
   &OPTIONAL; trailer section containing trailer fields. Chunked enables content
   streams of unknown size to be transferred as a sequence of length-delimited
   buffers, which enables the sender to retain connection persistence and the
   recipient to know when it has received the entire message.
</t>
<sourcecode type="abnf7230"><iref primary="true" item="Grammar" subitem="chunked-body"><!--terminal production--></iref><iref primary="true" item="Grammar" subitem="chunk"/><iref primary="true" item="Grammar" subitem="chunk-size"/><iref primary="true" item="Grammar" subitem="last-chunk"/><iref primary="false" item="Grammar" subitem="trailer-section"/><iref primary="false" item="Grammar" subitem="chunk-ext"/><iref primary="true" item="Grammar" subitem="chunk-data"/>
  <x:ref>chunked-body</x:ref>   = *<x:ref>chunk</x:ref>
                   <x:ref>last-chunk</x:ref>
                   <x:ref>trailer-section</x:ref>
                   <x:ref>CRLF</x:ref>

  <x:ref>chunk</x:ref>          = <x:ref>chunk-size</x:ref> [ <x:ref>chunk-ext</x:ref> ] <x:ref>CRLF</x:ref>
                   <x:ref>chunk-data</x:ref> <x:ref>CRLF</x:ref>
  <x:ref>chunk-size</x:ref>     = 1*<x:ref>HEXDIG</x:ref>
  <x:ref>last-chunk</x:ref>     = 1*("0") [ <x:ref>chunk-ext</x:ref> ] <x:ref>CRLF</x:ref>

  <x:ref>chunk-data</x:ref>     = 1*<x:ref>OCTET</x:ref> ; a sequence of chunk-size octets
</sourcecode>
<t>
   The chunk-size field is a string of hex digits indicating the size of
   the chunk-data in octets. The chunked transfer coding is complete when a
   chunk with a chunk-size of zero is received, possibly followed by a
   trailer section, and finally terminated by an empty line.
</t>
<t>
   A recipient &MUST; be able to parse and decode the chunked transfer coding.
</t>
<t>
   HTTP/1.1 does not define any means to limit the size of a
   chunked response such that an intermediary can be assured of buffering the
   entire response. Additionally, very large chunk sizes may cause overflows
   or loss of precision if their values are not represented accurately in a
   receiving implementation. Therefore, recipients &MUST; anticipate
   potentially large hexadecimal numerals and prevent parsing errors due to
   integer conversion overflows or precision loss due to integer
   representation.
</t>
<t>
   The chunked coding does not define any parameters. Their presence
   &SHOULD; be treated as an error.
</t>

<section title="Chunk Extensions" anchor="chunked.extension">
  <x:anchor-alias value="chunk-ext"/>
  <x:anchor-alias value="chunk-ext-name"/>
  <x:anchor-alias value="chunk-ext-val"/>
<t>
   The chunked coding allows each chunk to include zero or more chunk
   extensions, immediately following the <x:ref>chunk-size</x:ref>, for the
   sake of supplying per-chunk metadata (such as a signature or hash),
   mid-message control information, or randomization of message body size.
</t>
<sourcecode type="abnf7230"><iref primary="true" item="Grammar" subitem="chunk-ext"/><iref primary="true" item="Grammar" subitem="chunk-ext-name"/><iref primary="true" item="Grammar" subitem="chunk-ext-val"/>
  <x:ref>chunk-ext</x:ref>      = *( <x:ref>BWS</x:ref> ";" <x:ref>BWS</x:ref> <x:ref>chunk-ext-name</x:ref>
                      [ <x:ref>BWS</x:ref> "=" <x:ref>BWS</x:ref> <x:ref>chunk-ext-val</x:ref> ] )

  <x:ref>chunk-ext-name</x:ref> = <x:ref>token</x:ref>
  <x:ref>chunk-ext-val</x:ref>  = <x:ref>token</x:ref> / <x:ref>quoted-string</x:ref>
</sourcecode>
<t>
   The chunked coding is specific to each connection and is likely to be
   removed or recoded by each recipient (including intermediaries) before any
   higher-level application would have a chance to inspect the extensions.
   Hence, the use of chunk extensions is generally limited to specialized HTTP
   services such as "long polling" (where client and server can have shared
   expectations regarding the use of chunk extensions) or for padding within
   an end-to-end secured connection.
</t>
<t>
   A recipient &MUST; ignore unrecognized chunk extensions.
   A server ought to limit the total length of chunk extensions received in a
   request to an amount reasonable for the services provided, in the same way
   that it applies length limitations and timeouts for other parts of a
   message, and generate an appropriate <x:ref>4xx (Client Error)</x:ref>
   response if that amount is exceeded.
</t>
</section>

<section title="Chunked Trailer Section" anchor="chunked.trailer.section">
  <x:anchor-alias value="trailer-section"/>
<t>
   A trailer section allows the sender to include additional fields at the end
   of a chunked message in order to supply metadata that might be dynamically
   generated while the content is sent, such as a message integrity
   check, digital signature, or post-processing status. The proper use and
   limitations of trailer fields are defined in <xref target="trailer.fields"/>.
</t>
<sourcecode type="abnf7230"><iref primary="true" item="Grammar" subitem="trailer-section"/><iref primary="false" item="Grammar" subitem="field-line"/>
  <x:ref>trailer-section</x:ref>   = *( <x:ref>field-line</x:ref> <x:ref>CRLF</x:ref> )
</sourcecode>
<t>
   A recipient that removes the chunked coding from a message &MAY;
   selectively retain or discard the received trailer fields. A recipient
   that retains a received trailer field &MUST; either store/forward the
   trailer field separately from the received header fields or merge the
   received trailer field into the header section.
   A recipient &MUST-NOT; merge a received trailer field into the
   header section unless its corresponding header field definition
   explicitly permits and instructs how the trailer field value can be
   safely merged.
</t>
</section>

<section title="Decoding Chunked" anchor="decoding.chunked">
<t>
   A process for decoding the chunked transfer coding
   can be represented in pseudo-code as:
</t>
<sourcecode type="pseudocode">
  length := 0
  read chunk-size, chunk-ext (if any), and CRLF
  while (chunk-size &gt; 0) {
     read chunk-data and CRLF
     append chunk-data to content
     length := length + chunk-size
     read chunk-size, chunk-ext (if any), and CRLF
  }
  read trailer field
  while (trailer field is not empty) {
     if (trailer fields are stored/forwarded separately) {
         append trailer field to existing trailer fields
     }
     else if (trailer field is understood and defined as mergeable) {
         merge trailer field with existing header fields
     }
     else {
         discard trailer field
     }
     read trailer field
  }
  Content-Length := length
  Remove "chunked" from Transfer-Encoding
</sourcecode>
</section>
</section>

<section title="Transfer Codings for Compression" anchor="compression.codings">
  <iref primary="true" item="compress (transfer coding)"/>
  <iref primary="true" item="x-compress (transfer coding)"/>
  <iref primary="true" item="deflate (transfer coding)"/>
  <iref primary="true" item="gzip (transfer coding)"/>
  <iref primary="true" item="x-gzip (transfer coding)"/>
<t>
   The following transfer coding names for compression are defined by
   the same algorithm as their corresponding content coding:
</t>
<dl newline="true">
  <dt>compress (and x-compress)</dt>
  <dd>See <xref target="compress.coding"/>.</dd>
  <dt>deflate</dt>
  <dd>See <xref target="deflate.coding"/>.</dd>
  <dt>gzip (and x-gzip)</dt>
  <dd>See <xref target="gzip.coding"/>.</dd>
</dl>
<t>
   The compression codings do not define any parameters. The presence
   of parameters with any of these compression codings &SHOULD; be treated
   as an error.
</t>
</section>

<section title="Transfer Coding Registry" anchor="transfer.coding.registry">
<t>
   The "HTTP Transfer Coding Registry" defines the namespace for transfer
   coding names. It is maintained at <eref target="https://www.iana.org/assignments/http-parameters"/>.
</t>
<t>
   Registrations &MUST; include the following fields:
</t>
<ul>
  <li>Name</li>
  <li>Description</li>
  <li>Pointer to specification text</li>
</ul>
<t>
   Names of transfer codings &MUST-NOT; overlap with names of content codings
   (<xref target="content.codings"/>) unless the encoding transformation is identical, as
   is the case for the compression codings defined in
   <xref target="compression.codings"/>.
</t>
<t>
   The <x:ref>TE</x:ref> header field (<xref target="field.te"/>) uses a
   pseudo-parameter named "q" as the rank value when multiple transfer codings
   are acceptable. Future registrations of transfer codings &SHOULD-NOT;
   define parameters called "q" (case-insensitively) in order to avoid
   ambiguities.
</t>
<t>
   Values to be added to this namespace require IETF Review (see
   <xref target="RFC8126" x:fmt="of" x:sec="4.8"/>) and &MUST;
   conform to the purpose of transfer coding defined in this specification.
</t>
<t>
   Use of program names for the identification of encoding formats
   is not desirable and is discouraged for future encodings.
</t>
</section>

<section title="Negotiating Transfer Codings" anchor="transfer.coding.negotiation">
<t>
  The TE field (<xref target="field.te"/>) is used in HTTP/1.1 to indicate
  what transfer codings, besides chunked, the client is willing to accept
  in the response and whether the client is willing to preserve
  trailer fields in a chunked transfer coding.
</t>
<t>
   A client &MUST-NOT; send the chunked transfer coding name in TE;
   chunked is always acceptable for HTTP/1.1 recipients.
</t>
<t>
   Three examples of TE use are below.
</t>
<sourcecode type="http-message">
TE: deflate
TE:
TE: trailers, deflate;q=0.5
</sourcecode>
<t>
   When multiple transfer codings are acceptable, the client &MAY; rank the
   codings by preference using a case-insensitive "q" parameter (similar to
   the qvalues used in content negotiation fields; see <xref target="quality.values"/>). The rank value
   is a real number in the range 0 through 1, where 0.001 is the least
   preferred and 1 is the most preferred; a value of 0 means "not acceptable".
</t>
<t>
   If the TE field value is empty or if no TE field is present, the only
   acceptable transfer coding is chunked. A message with no transfer coding
   is always acceptable.
</t>
<t>
   The keyword "trailers" indicates that the sender will not discard trailer
   fields, as described in <xref target="trailer.fields"/>.
</t>
<t>
   Since the TE header field only applies to the immediate connection,
   a sender of TE &MUST; also send a "TE" connection option within the
   <x:ref>Connection</x:ref> header field (<xref target="field.connection"/>)
   in order to prevent the TE header field from being forwarded by intermediaries
   that do not support its semantics.
</t>
</section>
</section>

<section title="Handling Incomplete Messages" anchor="incomplete.messages">
<t>
   A server that receives an incomplete request message, usually due to a
   canceled request or a triggered timeout exception, &MAY; send an error
   response prior to closing the connection.
</t>
<t>
   A client that receives an incomplete response message, which can occur
   when a connection is closed prematurely or when decoding a supposedly
   chunked transfer coding fails, &MUST; record the message as incomplete.
   Cache requirements for incomplete responses are defined in
   <xref target="incomplete.responses"/>.
</t>
<t>
   If a response terminates in the middle of the header section (before the
   empty line is received) and the status code might rely on header fields to
   convey the full meaning of the response, then the client cannot assume
   that meaning has been conveyed; the client might need to repeat the
   request in order to determine what action to take next.
</t>
<t>
   A message body that uses the chunked transfer coding is
   incomplete if the zero-sized chunk that terminates the encoding has not
   been received.  A message that uses a valid <x:ref>Content-Length</x:ref> is
   incomplete if the size of the message body received (in octets) is less than
   the value given by Content-Length.  A response that has neither chunked
   transfer coding nor Content-Length is terminated by closure of the
   connection and, if the header section was received intact, is considered
   complete unless an error was indicated by the underlying connection
   (e.g., an "incomplete close" in TLS would leave the response incomplete,
   as described in <xref target="tls.connection.closure"/>).
</t>
</section>

<section title="Connection Management" anchor="connection.management">
<t>
   HTTP messaging is independent of the underlying transport- or
   session-layer connection protocol(s).  HTTP only presumes a reliable
   transport with in-order delivery of requests and the corresponding
   in-order delivery of responses.  The mapping of HTTP request and
   response structures onto the data units of an underlying transport
   protocol is outside the scope of this specification.
</t>
<t>
   As described in <xref target="routing.inbound"/>, the specific
   connection protocols to be used for an HTTP interaction are determined by
   client configuration and the <x:ref>target URI</x:ref>.
   For example, the "http" URI scheme
   (<xref target="http.uri"/>) indicates a default connection of TCP
   over IP, with a default TCP port of 80, but the client might be
   configured to use a proxy via some other connection, port, or protocol.
</t>
<t>
   HTTP implementations are expected to engage in connection management,
   which includes maintaining the state of current connections,
   establishing a new connection or reusing an existing connection,
   processing messages received on a connection, detecting connection
   failures, and closing each connection.
   Most clients maintain multiple connections in parallel, including
   more than one connection per server endpoint.
   Most servers are designed to maintain thousands of concurrent connections,
   while controlling request queues to enable fair use and detect
   denial-of-service attacks.
</t>

<section title="Establishment" anchor="persistent.establishment">
<t>
   It is beyond the scope of this specification to describe how connections
   are established via various transport- or session-layer protocols.
   Each HTTP connection maps to one underlying transport connection.
</t>
</section>

<section title="Associating a Response to a Request" anchor="associating.response.to.request">
<t>
   HTTP/1.1 does not include a request identifier for associating a given
   request message with its corresponding one or more response messages.
   Hence, it relies on the order of response arrival to correspond exactly
   to the order in which requests are made on the same connection.
   More than one response message per request only occurs when one or more
   informational responses (<x:ref>1xx</x:ref>; see <xref target="status.1xx"/>) precede a
   final response to the same request.
</t>
<t>
   A client that has more than one outstanding request on a connection &MUST;
   maintain a list of outstanding requests in the order sent and &MUST;
   associate each received response message on that connection to the
   first outstanding request that has not yet received a final
   (non-<x:ref>1xx</x:ref>) response.
</t>
<t>
   If a client receives data on a connection that doesn't have
   outstanding requests, the client &MUST-NOT; consider that data to be a
   valid response; the client &SHOULD; close the connection, since message
   delimitation is now ambiguous, unless the data consists only of one or
   more CRLF (which can be discarded per <xref target="message.parsing"/>).
</t>
</section>

<section title="Persistence" anchor="persistent.connections">
   <x:anchor-alias value="persistent connections"/>
   <iref primary="false" item="close" x:for-anchor=""/>
<t>
   HTTP/1.1 defaults to the use of <x:dfn>persistent connections</x:dfn>,
   allowing multiple requests and responses to be carried over a single
   connection. HTTP implementations &SHOULD; support persistent connections.
</t>
<t>
   A recipient determines whether a connection is persistent or not based on
   the protocol version and <x:ref>Connection</x:ref> header field
   (<xref target="field.connection"/>) in the
   most recently received message, if any:
</t>
<ul>
   <li>If the "<x:ref>close</x:ref>" connection option is present
       (<xref target="persistent.tear-down"/>), the
       connection will not persist after the current response; else,</li>
   <li>If the received protocol is HTTP/1.1 (or later), the connection will
       persist after the current response; else,</li>
   <li>If the received protocol is HTTP/1.0, the "keep-alive" connection
       option is present, either the recipient is not a proxy or the
       message is a response, and the recipient wishes to honor the
       HTTP/1.0 "keep-alive" mechanism, the connection will persist after
       the current response; otherwise,</li>
   <li>The connection will close after the current response.</li>
</ul>
<t>
   A client that does not support <x:ref>persistent connections</x:ref> &MUST;
   send the "<x:ref>close</x:ref>" connection option in every request message.
</t>
<t>
   A server that does not support <x:ref>persistent connections</x:ref> &MUST;
   send the "<x:ref>close</x:ref>" connection option in every response message
   that does not have a <x:ref>1xx (Informational)</x:ref> status code.
</t>
<t>
   A client &MAY; send additional requests on a persistent connection until it
   sends or receives a "<x:ref>close</x:ref>" connection option or receives an
   HTTP/1.0 response without a "keep-alive" connection option.
</t>
<t>
   In order to remain persistent, all messages on a connection need to
   have a self-defined message length (i.e., one not defined by closure
   of the connection), as described in <xref target="message.body"/>.
   A server &MUST; read the entire request message body or close
   the connection after sending its response; otherwise, the
   remaining data on a persistent connection would be misinterpreted
   as the next request.  Likewise,
   a client &MUST; read the entire response message body if it intends
   to reuse the same connection for a subsequent request.
</t>
<t>
   A proxy server &MUST-NOT; maintain a persistent connection with an
   HTTP/1.0 client (see <xref target="compatibility.with.http.1.0.persistent.connections"/> for
   information and discussion of the problems with the Keep-Alive header field
   implemented by many HTTP/1.0 clients).
</t>
<t>
   See <xref target="compatibility.with.http.1.0.persistent.connections"/>
   for more information on backwards compatibility with HTTP/1.0 clients.
</t>

<section title="Retrying Requests" anchor="persistent.retrying.requests">
<t>
   Connections can be closed at any time, with or without intention.
   Implementations ought to anticipate the need to recover
   from asynchronous close events. The conditions under which a client can
   automatically retry a sequence of outstanding requests are defined in
   <xref target="idempotent.methods"/>.
 </t>
</section>

<section title="Pipelining" anchor="pipelining">
   <x:anchor-alias value="pipeline"/>
<t>
   A client that supports persistent connections &MAY; <x:dfn>pipeline</x:dfn>
   its requests (i.e., send multiple requests without waiting for each
   response). A server &MAY; process a sequence of pipelined requests in
   parallel if they all have safe methods (<xref target="safe.methods"/>), but it &MUST; send
   the corresponding responses in the same order that the requests were
   received.
</t>
<t>
   A client that pipelines requests &SHOULD; retry unanswered requests if the
   connection closes before it receives all of the corresponding responses.
   When retrying pipelined requests after a failed connection (a connection
   not explicitly closed by the server in its last complete response), a
   client &MUST-NOT; pipeline immediately after connection establishment,
   since the first remaining request in the prior pipeline might have caused
   an error response that can be lost again if multiple requests are sent on a
   prematurely closed connection (see the TCP reset problem described in
   <xref target="persistent.tear-down"/>).
</t>
<t>
   Idempotent methods (<xref target="idempotent.methods"/>) are significant to pipelining
   because they can be automatically retried after a connection failure.
   A user agent &SHOULD-NOT; pipeline requests after a non-idempotent method,
   until the final response status code for that method has been received,
   unless the user agent has a means to detect and recover from partial
   failure conditions involving the pipelined sequence.
</t>
<t>
   An intermediary that receives pipelined requests &MAY; pipeline those
   requests when forwarding them inbound, since it can rely on the outbound
   user agent(s) to determine what requests can be safely pipelined. If the
   inbound connection fails before receiving a response, the pipelining
   intermediary &MAY; attempt to retry a sequence of requests that have yet
   to receive a response if the requests all have idempotent methods;
   otherwise, the pipelining intermediary &SHOULD; forward any received
   responses and then close the corresponding outbound connection(s) so that
   the outbound user agent(s) can recover accordingly.
</t>
</section>
</section>

<section title="Concurrency" anchor="persistent.concurrency">
<t>
   A client ought to limit the number of simultaneous open
   connections that it maintains to a given server.
</t>
<t>
   Previous revisions of HTTP gave a specific number of connections as a
   ceiling, but this was found to be impractical for many applications. As a
   result, this specification does not mandate a particular maximum number of
   connections but, instead, encourages clients to be conservative when opening
   multiple connections.
</t>
<t>
   Multiple connections are typically used to avoid the "head-of-line
   blocking" problem, wherein a request that takes significant server-side
   processing and/or transfers very large content would block subsequent
   requests on the
   same connection. However, each connection consumes server resources.
</t>
<t>
   Furthermore, using multiple connections can cause undesirable side effects
   in congested networks.
   Using larger numbers of connections can also cause side effects in
   otherwise uncongested networks, because their aggregate and initially
   synchronized sending behavior can cause congestion that would not have
   been present if fewer parallel connections had been used.
</t>
<t>
   Note that a server might reject traffic that it deems abusive or
   characteristic of a denial-of-service attack, such as an excessive number
   of open connections from a single client.
</t>
</section>

<section title="Failures and Timeouts" anchor="persistent.failures">
<t>
   Servers will usually have some timeout value beyond which they will
   no longer maintain an inactive connection. Proxy servers might make
   this a higher value since it is likely that the client will be making
   more connections through the same proxy server. The use of persistent
   connections places no requirements on the length (or existence) of
   this timeout for either the client or the server.
</t>
<t>
   A client or server that wishes to time out &SHOULD; issue a graceful close
   on the connection. Implementations &SHOULD; constantly monitor open
   connections for a received closure signal and respond to it as appropriate,
   since prompt closure of both sides of a connection enables allocated system
   resources to be reclaimed.
</t>
<t>
   A client, server, or proxy &MAY; close the transport connection at any
   time. For example, a client might have started to send a new request
   at the same time that the server has decided to close the "idle"
   connection. From the server's point of view, the connection is being
   closed while it was idle, but from the client's point of view, a
   request is in progress.
</t>
<t>
   A server &SHOULD; sustain persistent connections, when possible, and allow
   the underlying transport's flow-control mechanisms to resolve temporary overloads rather
   than terminate connections with the expectation that clients will retry.
   The latter technique can exacerbate network congestion or server load.
</t>
<t>
   A client sending a message body &SHOULD; monitor
   the network connection for an error response while it is transmitting
   the request. If the client sees a response that indicates the server does
   not wish to receive the message body and is closing the connection, the
   client &SHOULD; immediately cease transmitting the body and close its side
   of the connection.
</t>
</section>

<section title="Tear-down" anchor="persistent.tear-down">
  <iref primary="false" item="Connection header field" x:for-anchor=""/>
  <iref primary="true" item="close" x:for-anchor=""/>
  <x:anchor-alias value="close"/>
<t>
   The "close" connection option is defined as a signal that the sender
   will close this connection after completion of the response.
   A sender &SHOULD; send a <x:ref>Connection</x:ref> header field
   (<xref target="field.connection"/>) containing the "close" connection option
   when it intends to close a connection. For example,
</t>
<sourcecode type="http-message">
Connection: close
</sourcecode>
<t>
   as a request header field indicates that this is the last request that
   the client will send on this connection, while in a response, the same
   field indicates that the server is going to close this connection after
   the response message is complete.
</t>
<t anchor="field.close">
   <iref primary="true" item="Fields" subitem="Close" x:for-anchor=""/>
   <rdf:Description>
     <comments xmlns="urn:ietf:id:draft-ietf-httpbis-p2-semantics#">(reserved)</comments>
   </rdf:Description>
   Note that the field name "Close" is reserved, since using that name as a
   header field might conflict with the "close" connection option.
</t>
<t>
   A client that sends a "close" connection option &MUST-NOT;
   send further requests on that connection (after the one containing the
   "close") and &MUST; close the connection after reading the
   final response message corresponding to this request.
</t>
<t>
   A server that receives a "close" connection option &MUST;
   initiate closure of the connection (see below) after it sends the
   final response to the request that contained the "close" connection option.
   The server &SHOULD; send a "close" connection option in its final response
   on that connection. The server &MUST-NOT; process any further requests
   received on that connection.
</t>
<t>
   A server that sends a "close" connection option &MUST;
   initiate closure of the connection (see below) after it sends the
   response containing the "close" connection option. The server &MUST-NOT; process
   any further requests received on that connection.
</t>
<t>
   A client that receives a "close" connection option &MUST;
   cease sending requests on that connection and close the connection
   after reading the response message containing the "close" connection option;
   if additional pipelined requests had been sent on the connection,
   the client &SHOULD-NOT; assume that they will be processed by the server.
</t>
<t>
   If a server performs an immediate close of a TCP connection, there is a
   significant risk that the client will not be able to read the last HTTP
   response.  If the server receives additional data from the client on a
   fully closed connection, such as another request sent by the
   client before receiving the server's response, the server's TCP stack will
   send a reset packet to the client; unfortunately, the reset packet might
   erase the client's unacknowledged input buffers before they can be read
   and interpreted by the client's HTTP parser.
</t>
<t>
   To avoid the TCP reset problem, servers typically close a connection in
   stages. First, the server performs a half-close by closing only the write
   side of the read/write connection. The server then continues to read from
   the connection until it receives a corresponding close by the client, or
   until the server is reasonably certain that its own TCP stack has received
   the client's acknowledgement of the packet(s) containing the server's last
   response. Finally, the server fully closes the connection.
</t>
<t>
   It is unknown whether the reset problem is exclusive to TCP or might also
   be found in other transport connection protocols.
</t>
<t>
   Note that a TCP connection that is half-closed by the client does not
   delimit a request message, nor does it imply that the client is no longer
   interested in a response. In general, transport signals cannot be relied
   upon to signal edge cases, since HTTP/1.1 is independent of transport.
</t>
</section>

<section title="TLS Connection Initiation" anchor="tls.connection.initiation">
<t>
   Conceptually, HTTP/TLS is simply sending HTTP messages over a connection
   secured via TLS <xref target="TLS13"/>.
</t>
<t>
   The HTTP client also acts as the TLS client. It initiates a connection to
   the server on the appropriate port and sends the TLS ClientHello to begin
   the TLS handshake. When the TLS handshake has finished, the client may then
   initiate the first HTTP request. All HTTP data &MUST; be sent as TLS
   "application data" but is otherwise treated like a normal connection for
   HTTP (including potential reuse as a persistent connection).
</t>
</section>

<section title="TLS Connection Closure" anchor="tls.connection.closure">
<t>
   TLS uses an exchange of closure alerts prior to (non-error) connection
   closure to provide secure connection closure; see <xref section="6.1" target="TLS13"/>.  When a
   valid closure alert is received, an implementation can be assured that no
   further data will be received on that connection.
</t>
<t>
   When an implementation knows that it has sent or received all the
   message data that it cares about, typically by detecting HTTP message
   boundaries, it might generate an "incomplete close" by sending a
   closure alert and then closing the connection without waiting to
   receive the corresponding closure alert from its peer.
</t>
<t>
   An incomplete close does not call into question the security of the data
   already received, but it could indicate that subsequent data might have been
   truncated. As TLS is not directly aware of HTTP message framing, it is
   necessary to examine the HTTP data itself to determine whether messages are
   complete. Handling of incomplete messages is defined in
   <xref target="incomplete.messages"/>.
</t>
<t>
   When encountering an incomplete close, a client &SHOULD; treat as completed
   all requests for which it has received either
</t>
<ol>
  <li>
    as much data as specified in the <x:ref>Content-Length</x:ref> header
    field or
  </li>
  <li>
    the terminal zero-length chunk (when <x:ref>Transfer-Encoding</x:ref> of chunked is used).
  </li>
</ol>   
<t>
   A response that has neither chunked
   transfer coding nor Content-Length is complete only if a valid closure alert
   has been received. Treating an incomplete message as complete could expose
   implementations to attack.
</t>
<t>
   A client detecting an incomplete close &SHOULD; recover gracefully.
</t>
<t>
   Clients &MUST; send a closure alert before closing the connection.
   Clients that do not expect to receive any more data &MAY; choose not
   to wait for the server's closure alert and simply close the
   connection, thus generating an incomplete close on the server side.
</t>
<t>
   Servers &SHOULD; be prepared to receive an incomplete close from the client,
   since the client can often locate the end of server data.
</t>
<t>
   Servers &MUST; attempt to initiate an exchange of closure alerts with
   the client before closing the connection. Servers &MAY; close the
   connection after sending the closure alert, thus generating an
   incomplete close on the client side.
</t>
</section>
</section>

<section title="Enclosing Messages as Data" anchor="enclosing.messages">

<section title="Media Type message/http" anchor="media.type.message.http">
<iref item="Media Type" subitem="message/http" primary="true"/>
<iref item="message/http Media Type" primary="true"/>
<t>
   The "message/http" media type can be used to enclose a single HTTP request or
   response message, provided that it obeys the MIME restrictions for all
   "message" types regarding line length and encodings. Because of the line
   length limitations, field values within "message/http" are allowed to use
   line folding (<x:ref>obs-fold</x:ref>), as described in
   <xref target="line.folding"/>, to convey the field value over multiple
   lines. A recipient of "message/http" data &MUST; replace any obsolete line
   folding with one or more SP characters when the message is consumed.
</t>
<dl>
  <dt>Type name:</dt>
  <dd>message</dd>
  <dt>Subtype name:</dt>
  <dd>http</dd>
  <dt>Required parameters:</dt>
  <dd>N/A</dd>
  <dt>Optional parameters:</dt>
  <dd>
    <t>version, msgtype</t>
    <dl>
      <dt>version:</dt>
      <dd>
        The HTTP-version number of the enclosed message
        (e.g., "1.1"). If not present, the version can be
        determined from the first line of the body.
      </dd>
      <dt>msgtype:</dt>
      <dd>
        The message type &mdash; "request" or "response". If not
        present, the type can be determined from the first
        line of the body.
      </dd>
    </dl>
  </dd>
  <dt>Encoding considerations:</dt>
  <dd>only "7bit", "8bit", or "binary" are permitted</dd>
  <dt>Security considerations:</dt>
  <dd>see <xref target="security.considerations"/></dd>
  <dt>Interoperability considerations:</dt>
  <dd>N/A</dd>
  <dt>Published specification:</dt>
  <dd>RFC 9112 (see <xref target="media.type.message.http"/>).</dd>
  <dt>Applications that use this media type:</dt>
  <dd>N/A</dd>
  <dt>Fragment identifier considerations:</dt>
  <dd>N/A</dd>
  <dt>Additional information:</dt>
  <dd>
    <dl>
      <dt>Magic number(s):</dt>
      <dd>N/A</dd>
      <dt>Deprecated alias names for this type:</dt>
      <dd>N/A</dd>
      <dt>File extension(s):</dt>
      <dd>N/A</dd>
      <dt>Macintosh file type code(s):</dt>
      <dd>N/A</dd>
    </dl>
  </dd>
  <dt>Person and email address to contact for further information:</dt>
  <dd>See Authors' Addresses section.</dd>
  <dt>Intended usage:</dt>
  <dd>COMMON</dd>
  <dt>Restrictions on usage:</dt>
  <dd>N/A</dd>
  <dt>Author:</dt>
  <dd>See Authors' Addresses section.</dd>
  <dt>Change controller:</dt>
  <dd>IESG</dd>
</dl>
</section>

<section title="Media Type application/http" anchor="media.type.application.http">
<iref item="Media Type" subitem="application/http" primary="true"/>
<iref item="application/http Media Type" primary="true"/>
<t>
   The "application/http" media type can be used to enclose a pipeline of one or more
   HTTP request or response messages (not intermixed).
</t>
<dl>
  <dt>Type name:</dt>
  <dd>application</dd>
  <dt>Subtype name:</dt>
  <dd>http</dd>
  <dt>Required parameters:</dt>
  <dd>N/A</dd>
  <dt>Optional parameters:</dt>
  <dd>
    <t>
      version, msgtype
    </t>
    <dl>
      <dt>version:</dt>
      <dd>
        The HTTP-version number of the enclosed messages
        (e.g., "1.1"). If not present, the version can be
        determined from the first line of the body.
      </dd>
      <dt>msgtype:</dt>
      <dd>
        The message type &mdash; "request" or "response". If not
        present, the type can be determined from the first
        line of the body.
      </dd>
    </dl>
  </dd>
  <dt>Encoding considerations:</dt>
  <dd>
    HTTP messages enclosed by this type
    are in "binary" format; use of an appropriate
    Content-Transfer-Encoding is required when
    transmitted via email.
  </dd>
  <dt>Security considerations:</dt>
  <dd>
    see <xref target="security.considerations"/>
  </dd>
  <dt>Interoperability considerations:</dt>
  <dd>N/A</dd>
  <dt>Published specification:</dt>
  <dd>
    RFC 9112 (see <xref target="media.type.application.http"/>).
  </dd>
  <dt>Applications that use this media type:</dt>
  <dd>N/A</dd>
  <dt>Fragment identifier considerations:</dt>
  <dd>N/A</dd>
  <dt>Additional information:</dt>
  <dd>
    <dl>
      <dt>Deprecated alias names for this type:</dt>
      <dd>N/A</dd>
      <dt>Magic number(s):</dt>
      <dd>N/A</dd>
      <dt>File extension(s):</dt>
      <dd>N/A</dd>
      <dt>Macintosh file type code(s):</dt>
      <dd>N/A</dd>
    </dl>
  </dd>
  <dt>Person and email address to contact for further information:</dt>
  <dd>See Authors' Addresses section.</dd>
  <dt>Intended usage:</dt>
  <dd>COMMON</dd>
  <dt>Restrictions on usage:</dt>
  <dd>N/A</dd>
  <dt>Author:</dt>
  <dd>See Authors' Addresses section.</dd>
  <dt>Change controller:</dt>
  <dd>IESG</dd>
</dl>
</section>
</section>

<section title="Security Considerations" anchor="security.considerations">
<t>
   This section is meant to inform developers, information providers, and
   users about known security considerations relevant to HTTP message syntax
   and parsing. Security considerations about HTTP semantics,
   content, and routing are addressed in <xref target="HTTP"/>.
</t>

<section title="Response Splitting" anchor="response.splitting">
<t>
   Response splitting (a.k.a. CRLF injection) is a common technique, used in
   various attacks on Web usage, that exploits the line-based nature of HTTP
   message framing and the ordered association of requests to responses on
   persistent connections <xref target="Klein"/>. This technique can be
   particularly damaging when the requests pass through a shared cache.
</t>
<t>
   Response splitting exploits a vulnerability in servers (usually within an
   application server) where an attacker can send encoded data within some
   parameter of the request that is later decoded and echoed within any of the
   response header fields of the response. If the decoded data is crafted to
   look like the response has ended and a subsequent response has begun, the
   response has been split, and the content within the apparent second response
   is controlled by the attacker. The attacker can then make any other request
   on the same persistent connection and trick the recipients (including
   intermediaries) into believing that the second half of the split is an
   authoritative answer to the second request.
</t>
<t>
   For example, a parameter within the request-target might be read by an
   application server and reused within a redirect, resulting in the same
   parameter being echoed in the <x:ref>Location</x:ref> header field of the
   response. If the parameter is decoded by the application and not properly
   encoded when placed in the response field, the attacker can send encoded
   CRLF octets and other content that will make the application's single
   response look like two or more responses.
</t>
<t>
   A common defense against response splitting is to filter requests for data
   that looks like encoded CR and LF (e.g., "%0D" and "%0A"). However, that
   assumes the application server is only performing URI decoding rather
   than more obscure data transformations like charset transcoding, XML entity
   translation, base64 decoding, sprintf reformatting, etc.  A more effective
   mitigation is to prevent anything other than the server's core protocol
   libraries from sending a CR or LF within the header section, which means
   restricting the output of header fields to APIs that filter for bad octets
   and not allowing application servers to write directly to the protocol
   stream.
</t>
</section>

<section title="Request Smuggling" anchor="request.smuggling">
<t>
   Request smuggling (<xref target="Linhart"/>) is a technique that exploits
   differences in protocol parsing among various recipients to hide additional
   requests (which might otherwise be blocked or disabled by policy) within an
   apparently harmless request.  Like response splitting, request smuggling
   can lead to a variety of attacks on HTTP usage.
</t>
<t>
   This specification has introduced new requirements on request parsing,
   particularly with regard to message framing in
   <xref target="message.body.length"/>, to reduce the effectiveness of
   request smuggling.
</t>
</section>

<section title="Message Integrity" anchor="message.integrity">
<t>
   HTTP does not define a specific mechanism for ensuring message integrity,
   instead relying on the error-detection ability of underlying transport
   protocols and the use of length or chunk-delimited framing to detect
   completeness. Historically, the lack of
   a single integrity mechanism has been justified by the informal nature of
   most HTTP communication.  However, the prevalence of HTTP as an information
   access mechanism has resulted in its increasing use within environments
   where verification of message integrity is crucial.
</t>
<t>
   The mechanisms provided with the "https" scheme, such as authenticated
   encryption, provide protection against modification of messages. Care
   is needed, however, to ensure that connection closure cannot be used to
   truncate messages (see <xref target="tls.connection.closure"/>). User agents
   might refuse to accept incomplete messages or treat them specially. For
   example, a browser being used to view medical history or drug interaction
   information needs to indicate to the user when such information is detected
   by the protocol to be incomplete, expired, or corrupted during transfer.
   Such mechanisms might be selectively enabled via user agent extensions or
   the presence of message integrity metadata in a response.
</t>
<t>
   The "http" scheme provides no protection against accidental or malicious
   modification of messages.
</t>
<t>
   Extensions to the protocol might be used to mitigate the risk of unwanted
   modification of messages by intermediaries, even when the "https" scheme is
   used. Integrity might be assured by using message authentication codes
   or digital
   signatures that are selectively added to messages via extensible metadata
   fields.
</t>
</section>

<section title="Message Confidentiality" anchor="message.confidentiality">
<t>
   HTTP relies on underlying transport protocols to provide message
   confidentiality when that is desired. HTTP has been specifically designed
   to be independent of the transport protocol, such that it can be used
   over many forms of encrypted connection, with the selection of
   such transports being identified by the choice of URI scheme or within
   user agent configuration.
</t>
<t>
   The "https" scheme can be used to identify resources that require a
   confidential connection, as described in <xref target="https.uri"/>.
</t>
</section>
</section>

<section title="IANA Considerations" anchor="IANA.considerations">
<t>
   The change controller for the following registrations is:
   "IETF (iesg@ietf.org) - Internet Engineering Task Force".
</t>

<section title="Field Name Registration" anchor="field.name.registration">
<t>
   IANA has added the following field names to the "Hypertext Transfer Protocol (HTTP) Field
   Name Registry" at <eref target="https://www.iana.org/assignments/http-fields"/>,
   as described in <xref target="HTTP" x:rel="#field.name.registration"/>.
</t>
<?BEGININC build/draft-ietf-httpbis-messaging-latest.iana-headers ?>
<!--AUTOGENERATED FROM extract-header-defs.xslt, do not edit manually-->
<table align="left" anchor="iana.header.registration.table">
   <thead>
      <tr>
         <th>Field Name</th>
         <th>Status</th>
         <th>Section</th>
         <th>Comments</th>
      </tr>
   </thead>
   <tbody>
      <tr>
         <td>Close</td>
         <td>permanent</td>
         <td>
            <xref target="persistent.tear-down" format="counter"/>
         </td>
         <td>(reserved)</td>
      </tr>
      <tr>
         <td>MIME-Version</td>
         <td>permanent</td>
         <td>
            <xref target="mime-version" format="counter"/>
         </td>
         <td/>
      </tr>
      <tr>
         <td>Transfer-Encoding</td>
         <td>permanent</td>
         <td>
            <xref target="field.transfer-encoding" format="counter"/>
         </td>
         <td/>
      </tr>
   </tbody>
</table>
<!--(END)-->
<?ENDINC build/draft-ietf-httpbis-messaging-latest.iana-headers ?>
</section>

<section title="Media Type Registration" anchor="media.type.http">
<t>
   IANA has updated the "Media Types" registry at
   <eref target="https://www.iana.org/assignments/media-types"/>
   with the registration information in Sections
   <xref target="media.type.message.http" format="counter"/> and
   <xref target="media.type.application.http" format="counter"/> for the media types
   "message/http" and "application/http", respectively.
</t>
</section>

<section title="Transfer Coding Registration" anchor="transfer.coding.registration">
<t>
   IANA has updated the "HTTP Transfer Coding Registry" at
   <eref target="https://www.iana.org/assignments/http-parameters/"/>
   with the registration procedure of <xref target="transfer.coding.registry"/>
   and the content coding names summarized in the table below.
</t>
<table align="left" anchor="iana.transfer.coding.registration.table">
  <thead>
    <tr>
      <th>Name</th>
      <th>Description</th>
      <th>Section</th>
    </tr>
  </thead>
  <tbody>
    <tr>
      <td>chunked</td>
      <td>Transfer in a series of chunks</td>
      <td><xref target="chunked.encoding" format="counter"/></td>
    </tr>
    <tr>
      <td>compress</td>
      <td>UNIX "compress" data format <xref target="Welch"/></td>
      <td><xref target="compression.codings" format="counter"/></td>
    </tr>
    <tr>
      <td>deflate</td>
      <td>"deflate" compressed data (<xref target="RFC1951"/>) inside
      the "zlib" data format (<xref target="RFC1950"/>)</td>
      <td><xref target="compression.codings" format="counter"/></td>
    </tr>
    <tr>
      <td>gzip</td>
      <td>GZIP file format <xref target="RFC1952"/></td>
      <td><xref target="compression.codings" format="counter"/></td>
    </tr>
    <tr>
      <td>trailers</td>
      <td>(reserved)</td>
      <td><xref target="transfer.coding.registration" format="counter"/></td>
    </tr>
    <tr>
      <td>x-compress</td>
      <td>Deprecated (alias for compress)</td>
      <td><xref target="compression.codings" format="counter"/></td>
    </tr>
    <tr>
      <td>x-gzip</td>
      <td>Deprecated (alias for gzip)</td>
      <td><xref target="compression.codings" format="counter"/></td>
    </tr>
  </tbody>
</table>
<aside>
  <t>
    &Note; the coding name "trailers" is reserved because its use would
    conflict with the keyword "trailers" in the <x:ref>TE</x:ref>
    header field (<xref target="field.te"/>).
  </t>
</aside>
</section>

<section title="ALPN Protocol ID Registration" anchor="alpn.registration">
<t>
   IANA has updated the
   "TLS Application-Layer Protocol Negotiation (ALPN) Protocol IDs" registry at
   <eref target="https://www.iana.org/assignments/tls-extensiontype-values/"/>
   with the registration below:
</t>
<table>
  <thead>
    <tr>
      <th>Protocol</th>
      <th>Identification Sequence</th>
      <th>Reference</th>
    </tr>
  </thead>
  <tbody>
    <tr>
      <td>HTTP/1.1</td>
      <td>0x68 0x74 0x74 0x70 0x2f 0x31 0x2e 0x31 ("http/1.1")</td>
      <td>RFC 9112</td>
    </tr>
  </tbody>
</table>
</section>

</section>

</middle>
<back>
<displayreference target="HTTP10" to="HTTP/1.0"/>

<references title="Normative References">

<reference anchor="HTTP">
  <x:source href="rfc9110.xml" basename="rfc9110">
    <x:defines>OWS</x:defines>
    <x:defines>RWS</x:defines>
    <x:defines>BWS</x:defines>
    <x:defines>port</x:defines>
    <x:defines>transfer-coding</x:defines>
    <x:defines>uri-host</x:defines>
    <x:has anchor="CONNECT"/>
    <x:has anchor="OPTIONS"/>
    <x:has anchor="abnf.extension"/>
    <x:has anchor="routing.inbound"/>
    <x:has anchor="routing.reject"/>
    <x:has anchor="content.codings"/>
    <x:has anchor="content.semantics"/>
    <x:has anchor="fields.names"/>
    <x:has anchor="fields.values"/>
    <x:has anchor="field.connection"/>
    <x:has anchor="field.content-length"/>
    <x:has anchor="field.host"/>
    <x:has anchor="field.te"/>
    <x:has anchor="http.date"/>
    <x:has anchor="http.uri"/>
    <x:has anchor="https.uri"/>
    <x:has anchor="idempotent.methods"/>
    <x:has anchor="message.forwarding"/>
    <x:has anchor="methods"/>
    <x:has anchor="multipart.byteranges"/>
    <x:has anchor="content"/>
    <x:has anchor="protocol.version"/>
    <x:has anchor="quality.values"/>
    <x:has anchor="quoted.strings"/>
    <x:has anchor="safe.methods"/>
    <x:has anchor="status.1xx"/>
    <x:has anchor="status.304"/>
    <x:has anchor="status.414"/>
    <x:has anchor="status.codes"/>
    <x:has anchor="tokens"/>
    <x:has anchor="target.resource"/>
    <x:has anchor="trailer.fields"/>
    <x:has anchor="uri.references"/>
    <x:has anchor="whitespace"/>
    <x:has anchor="compress.coding"/>
    <x:has anchor="deflate.coding"/>
    <x:has anchor="gzip.coding"/>
    <x:has anchor="uri.schemes"/>
  </x:source>
</reference>

<reference anchor="CACHING">
  <x:source href="rfc9111.xml" basename="rfc9111">
    <x:has anchor="incomplete.responses"/>
  </x:source>
</reference>

<reference anchor="URI">
 <front>
  <title abbrev='URI Generic Syntax'>Uniform Resource Identifier (URI): Generic Syntax</title>
  <author initials='T.' surname='Berners-Lee' fullname='Tim Berners-Lee'/>
  <author initials='R.' surname='Fielding' fullname='Roy T. Fielding'/>
  <author initials='L.' surname='Masinter' fullname='Larry Masinter'/>
  <date month='January' year='2005'></date>
 </front>
 <seriesInfo name="STD" value="66"/>
 <seriesInfo name="RFC" value="3986"/>
</reference>

<reference anchor="RFC5234">
  <front>
    <title abbrev="ABNF for Syntax Specifications">Augmented BNF for Syntax Specifications: ABNF</title>
    <author initials="D." surname="Crocker" fullname="Dave Crocker" role="editor"/>
    <author initials="P." surname="Overell" fullname="Paul Overell"/>
    <date month="January" year="2008"/>
  </front>
  <seriesInfo name="STD" value="68"/>
  <seriesInfo name="RFC" value="5234"/>
</reference>

<reference anchor="RFC7405">
  <front>
    <title>Case-Sensitive String Support in ABNF</title>
    <author initials="P." surname="Kyzivat" fullname="Dave Kyzivat"/>
    <date month="December" year="2014"/>
  </front>
  <seriesInfo name="RFC" value="7405"/>
</reference>

<reference anchor="RFC2119">
  <front>
    <title>Key words for use in RFCs to Indicate Requirement Levels</title>
    <author initials="S." surname="Bradner" fullname="Scott Bradner"/>
    <date month="March" year="1997"/>
  </front>
  <seriesInfo name="BCP" value="14"/>
  <seriesInfo name="RFC" value="2119"/>
</reference>

<reference anchor="RFC8174">
  <front>
    <title>Ambiguity of Uppercase vs Lowercase in RFC 2119 Key Words</title>
    <author initials='B.' surname='Leiba' fullname='Barry Leiba'/>
    <date year="2017" month="May"/>
  </front>
  <seriesInfo name='BCP' value='14'/>
  <seriesInfo name='RFC' value='8174'/>
</reference>

<reference anchor="USASCII">
  <front>
    <title>Coded Character Set -- 7-bit American Standard Code for Information Interchange</title>
    <author>
      <organization>American National Standards Institute</organization>
    </author>
    <date year="1986"/>
  </front>
  <seriesInfo name="ANSI" value="X3.4"/>
</reference>

<reference anchor="RFC1950">
  <front>
    <title>ZLIB Compressed Data Format Specification version 3.3</title>
    <author initials="P." surname="Deutsch" fullname="L. Peter Deutsch"/>
    <author initials="J-L." surname="Gailly" fullname="Jean-Loup Gailly"/>
    <date month="May" year="1996"/>
  </front>
  <seriesInfo name="RFC" value="1950"/>
</reference>

<reference anchor="RFC1951">
  <front>
    <title>DEFLATE Compressed Data Format Specification version 1.3</title>
    <author initials="P." surname="Deutsch" fullname="L. Peter Deutsch"/>
    <date month="May" year="1996"/>
  </front>
  <seriesInfo name="RFC" value="1951"/>
</reference>

<reference anchor="RFC1952">
  <front>
    <title>GZIP file format specification version 4.3</title>
    <author initials="P." surname="Deutsch" fullname="L. Peter Deutsch"/>
    <date month="May" year="1996"/>
  </front>
  <seriesInfo name="RFC" value="1952"/>
</reference>

<reference anchor="TLS13">
   <front>
      <title>The Transport Layer Security (TLS) Protocol Version 1.3</title>
      <author initials="E." surname="Rescorla" fullname="E. Rescorla"/>
      <date year="2018" month="August"/>
   </front>
   <seriesInfo name="RFC" value="8446"/>
</reference>

<reference anchor="Welch" target="https://ieeexplore.ieee.org/document/1659158/">
  <front>
    <title>A Technique for High-Performance Data Compression</title>
    <author initials="T." surname="Welch" fullname="Terry A. Welch"/>
    <date month="June" year="1984"/>
  </front>
  <seriesInfo name="IEEE Computer" value="17(6)"/>
  <seriesInfo name="DOI" value="10.1109/MC.1984.1659158"/>
</reference>

</references>

<references title="Informative References" anchor="informative.references">

<reference anchor="HTTP10">
  <front>
    <title abbrev="HTTP/1.0">Hypertext Transfer Protocol -- HTTP/1.0</title>
    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee"/>
    <author initials="R." surname="Fielding" fullname="Roy T. Fielding"/>
    <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen"/>
    <date month="May" year="1996"/>
  </front>
  <seriesInfo name="RFC" value="1945"/>
</reference>

<reference anchor="RFC2045">
  <front>
    <title abbrev="Internet Message Bodies">Multipurpose Internet Mail Extensions (MIME) Part One: Format of Internet Message Bodies</title>
    <author initials="N." surname="Freed" fullname="Ned Freed"/>
    <author initials="N." surname="Borenstein" fullname="Nathaniel S. Borenstein"/>
    <date month="November" year="1996"/>
  </front>
  <seriesInfo name="RFC" value="2045"/>
</reference>

<reference anchor="RFC2046">
  <front>
    <title abbrev="Media Types">Multipurpose Internet Mail Extensions (MIME) Part Two: Media Types</title>
    <author initials="N." surname="Freed" fullname="Ned Freed"/>
    <author initials="N." surname="Borenstein" fullname="Nathaniel S. Borenstein"/>
    <date month="November" year="1996"/>
  </front>
  <seriesInfo name="RFC" value="2046"/>
</reference>

<reference anchor="RFC2049">
  <front>
    <title abbrev="MIME Conformance">Multipurpose Internet Mail Extensions (MIME) Part Five: Conformance Criteria and Examples</title>
    <author initials="N." surname="Freed" fullname="Ned Freed"/>
    <author initials="N." surname="Borenstein" fullname="Nathaniel S. Borenstein"/>
    <date month="November" year="1996"/>
  </front>
  <seriesInfo name="RFC" value="2049"/>
</reference>

<reference anchor="RFC2068">
  <front>
    <title>Hypertext Transfer Protocol -- HTTP/1.1</title>
    <author initials="R." surname="Fielding" fullname="Roy T. Fielding"/>
    <author initials="J." surname="Gettys" fullname="Jim Gettys"/>
    <author initials="J." surname="Mogul" fullname="Jeffrey C. Mogul"/>
    <author initials="H." surname="Frystyk" fullname="Henrik Frystyk Nielsen"/>
    <author initials="T." surname="Berners-Lee" fullname="Tim Berners-Lee"/>
    <date month="January" year="1997"/>
  </front>
  <seriesInfo name="RFC" value="2068"/>
</reference>

<reference anchor="RFC2557">
  <front>
    <title abbrev="MIME Encapsulation of Aggregate Documents">MIME Encapsulation of Aggregate Documents, such as HTML (MHTML)</title>
    <author initials="J." surname="Palme" fullname="Jacob Palme"/>
    <author initials="A." surname="Hopmann" fullname="Alex Hopmann"/>
    <author initials="N." surname="Shelness" fullname="Nick Shelness"/>
    <date year="1999" month="March"/>
  </front>
  <seriesInfo name="RFC" value="2557"/>
</reference>

<reference anchor="RFC5322">
  <front>
    <title>Internet Message Format</title>
    <author initials="P." surname="Resnick" fullname="P. Resnick" role="editor"/>
    <date year="2008" month="October"/>
  </front>
  <seriesInfo name="RFC" value="5322"/>
</reference>

<reference anchor="RFC7230">
  <front>
    <title>Hypertext Transfer Protocol (HTTP/1.1): Message Syntax and Routing</title>
    <author initials="R." surname="Fielding" fullname="Roy T. Fielding" role="editor"/>
    <author initials="J." surname="Reschke" fullname="Julian F. Reschke" role="editor"/>
    <date month="June" year="2014"/>
  </front>
  <seriesInfo name="RFC" value="7230"/>
</reference>

<reference anchor="RFC8126">
  <front>
    <title>Guidelines for Writing an IANA Considerations Section in RFCs</title>
    <author initials="M." surname="Cotton" fullname="M. Cotton"/>
    <author initials="B." surname="Leiba" fullname="B. Leiba"/>
    <author initials="T." surname="Narten" fullname="T. Narten"/>
    <date year="2017" month="June" />
  </front>
  <seriesInfo name="BCP" value="26"/>
  <seriesInfo name="RFC" value="8126"/>
</reference>

<reference anchor="Klein" target="https://packetstormsecurity.com/papers/general/whitepaper_httpresponse.pdf">
  <front>
    <title>Divide and Conquer - HTTP Response Splitting, Web Cache Poisoning Attacks, and Related Topics</title>
    <author initials="A." surname="Klein" fullname="Amit Klein"/>
    <date year="2004" month="March"/>
  </front>
</reference>

<reference anchor="Linhart" target="https://www.cgisecurity.com/lib/HTTP-Request-Smuggling.pdf">
  <front>
    <title>HTTP Request Smuggling</title>
    <author initials="C." surname="Linhart" fullname="Chaim Linhart"/>
    <author initials="A." surname="Klein" fullname="Amit Klein"/>
    <author initials="R." surname="Heled" fullname="Ronen Heled"/>
    <author initials="S." surname="Orrin" fullname="Steve Orrin"/>
    <date year="2005" month="June"/>
  </front>
</reference>

</references>

<?BEGININC build/draft-ietf-httpbis-messaging-latest.abnf-appendix ?>
<section xmlns:x="http://purl.org/net/xml2rfc/ext" title="Collected ABNF" anchor="collected.abnf"><t>In the collected ABNF below, list rules are expanded per <xref target="HTTP" x:rel="#abnf.extension"/>.</t><sourcecode type="abnf" name="rfc9112.parsed-abnf">
<x:ref>BWS</x:ref> = &lt;BWS, see <xref target="HTTP" x:fmt="," x:sec="5.6.3"/>&gt;

<x:ref>HTTP-message</x:ref> = start-line CRLF *( field-line CRLF ) CRLF [
 message-body ]
<x:ref>HTTP-name</x:ref> = %x48.54.54.50 ; HTTP
<x:ref>HTTP-version</x:ref> = HTTP-name "/" DIGIT "." DIGIT

<x:ref>OWS</x:ref> = &lt;OWS, see <xref target="HTTP" x:fmt="," x:sec="5.6.3"/>&gt;

<x:ref>RWS</x:ref> = &lt;RWS, see <xref target="HTTP" x:fmt="," x:sec="5.6.3"/>&gt;

<x:ref>Transfer-Encoding</x:ref> = [ transfer-coding *( OWS "," OWS transfer-coding
 ) ]

<x:ref>absolute-URI</x:ref> = &lt;absolute-URI, see <xref target="URI" x:fmt="," x:sec="4.3"/>&gt;
<x:ref>absolute-form</x:ref> = absolute-URI
<x:ref>absolute-path</x:ref> = &lt;absolute-path, see <xref target="HTTP" x:fmt="," x:sec="4.1"/>&gt;
<x:ref>asterisk-form</x:ref> = "*"
<x:ref>authority</x:ref> = &lt;authority, see <xref target="URI" x:fmt="," x:sec="3.2"/>&gt;
<x:ref>authority-form</x:ref> = uri-host ":" port

<x:ref>chunk</x:ref> = chunk-size [ chunk-ext ] CRLF chunk-data CRLF
<x:ref>chunk-data</x:ref> = 1*OCTET
<x:ref>chunk-ext</x:ref> = *( BWS ";" BWS chunk-ext-name [ BWS "=" BWS chunk-ext-val
 ] )
<x:ref>chunk-ext-name</x:ref> = token
<x:ref>chunk-ext-val</x:ref> = token / quoted-string
<x:ref>chunk-size</x:ref> = 1*HEXDIG
<x:ref>chunked-body</x:ref> = *chunk last-chunk trailer-section CRLF

<x:ref>field-line</x:ref> = field-name ":" OWS field-value OWS
<x:ref>field-name</x:ref> = &lt;field-name, see <xref target="HTTP" x:fmt="," x:sec="5.1"/>&gt;
<x:ref>field-value</x:ref> = &lt;field-value, see <xref target="HTTP" x:fmt="," x:sec="5.5"/>&gt;

<x:ref>last-chunk</x:ref> = 1*"0" [ chunk-ext ] CRLF

<x:ref>message-body</x:ref> = *OCTET
<x:ref>method</x:ref> = token

<x:ref>obs-fold</x:ref> = OWS CRLF RWS
<x:ref>obs-text</x:ref> = &lt;obs-text, see <xref target="HTTP" x:fmt="," x:sec="5.6.4"/>&gt;
<x:ref>origin-form</x:ref> = absolute-path [ "?" query ]

<x:ref>port</x:ref> = &lt;port, see <xref target="URI" x:fmt="," x:sec="3.2.3"/>&gt;

<x:ref>query</x:ref> = &lt;query, see <xref target="URI" x:fmt="," x:sec="3.4"/>&gt;
<x:ref>quoted-string</x:ref> = &lt;quoted-string, see <xref target="HTTP" x:fmt="," x:sec="5.6.4"/>&gt;

<x:ref>reason-phrase</x:ref> = 1*( HTAB / SP / VCHAR / obs-text )
<x:ref>request-line</x:ref> = method SP request-target SP HTTP-version
<x:ref>request-target</x:ref> = origin-form / absolute-form / authority-form /
 asterisk-form

<x:ref>start-line</x:ref> = request-line / status-line
<x:ref>status-code</x:ref> = 3DIGIT
<x:ref>status-line</x:ref> = HTTP-version SP status-code SP [ reason-phrase ]

<x:ref>token</x:ref> = &lt;token, see <xref target="HTTP" x:fmt="," x:sec="5.6.2"/>&gt;
<x:ref>trailer-section</x:ref> = *( field-line CRLF )
<x:ref>transfer-coding</x:ref> = &lt;transfer-coding, see <xref target="HTTP" x:fmt="," x:sec="10.1.4"/>&gt;

<x:ref>uri-host</x:ref> = &lt;host, see <xref target="URI" x:fmt="," x:sec="3.2.2"/>&gt;
</sourcecode>
</section>
<?ENDINC build/draft-ietf-httpbis-messaging-latest.abnf-appendix ?>

<section title="Differences between HTTP and MIME" anchor="differences.between.http.and.mime">
<t>
   HTTP/1.1 uses many of the constructs defined for the Internet Message
   Format <xref target="RFC5322"/> and Multipurpose Internet Mail Extensions (MIME)
   <xref target="RFC2045"/> to allow a message body to be transmitted in an open
   variety of representations and with extensible fields.  However, some
   of these constructs have been reinterpreted to better fit the needs
   of interactive communication, leading to some differences in how MIME
   constructs are used within HTTP. These differences were carefully
   chosen to optimize performance over binary connections, allow
   greater freedom in the use of new media types, ease date comparisons,
   and accommodate common implementations.
</t>
<t>
   This appendix describes specific areas where HTTP differs from MIME.
   Proxies and gateways to and from strict MIME environments need to be
   aware of these differences and provide the appropriate conversions
   where necessary.
</t>

<section title="MIME-Version" anchor="mime-version">
  <iref primary="true" item="Fields" subitem="MIME-Version" x:for-anchor=""/><iref primary="true" item="Header Fields" subitem="MIME-Version" x:for-anchor=""/><iref primary="true" item="MIME-Version header field" x:for-anchor=""/>
  <x:anchor-alias value="MIME-Version"/>
<t>
   HTTP is not a MIME-compliant protocol. However, messages can
   include a single MIME-Version header field to indicate what
   version of the MIME protocol was used to construct the message. Use
   of the MIME-Version header field indicates that the message is in
   full conformance with the MIME protocol (as defined in <xref target="RFC2045"/>).
   Senders are responsible for ensuring full conformance (where
   possible) when exporting HTTP messages to strict MIME environments.
</t>
</section>

<section title="Conversion to Canonical Form" anchor="conversion.to.canonical.form">
<t>
   MIME requires that an Internet mail body part be converted to canonical
   form prior to being transferred, as described in <xref target="RFC2049"
   x:fmt="of" x:sec="4"/>, and that content with a type of "text" represents
   line breaks as CRLF, forbidding the use of CR or LF outside of line break
   sequences <xref target="RFC2046"/>. In contrast, HTTP does not care whether
   CRLF, bare CR, or bare LF are used to indicate a line break within content.
</t>
<t>
   A proxy or gateway from HTTP to a strict MIME
   environment ought to translate all line breaks within text media
   types to the RFC 2049 canonical form of CRLF. Note, however,
   this might be complicated by the presence of a <x:ref>Content-Encoding</x:ref>
   and by the fact that HTTP allows the use of some charsets
   that do not use octets 13 and 10 to represent CR and LF, respectively.
</t>
<t>
   Conversion will break any cryptographic
   checksums applied to the original content unless the original content
   is already in canonical form. Therefore, the canonical form is
   recommended for any content that uses such checksums in HTTP.
</t>
</section>

<section title="Conversion of Date Formats" anchor="conversion.of.date.formats">
<t>
   HTTP/1.1 uses a restricted set of date formats (<xref target="http.date"/>) to
   simplify the process of date comparison. Proxies and gateways from
   other protocols ought to ensure that any <x:ref>Date</x:ref> header field
   present in a message conforms to one of the HTTP/1.1 formats and rewrite
   the date if necessary.
</t>
</section>

<section title="Conversion of Content-Encoding" anchor="conversion.of.content-encoding">
<t>
   MIME does not include any concept equivalent to HTTP's
   <x:ref>Content-Encoding</x:ref> header field. Since this acts as a modifier
   on the media type, proxies and gateways from HTTP to MIME-compliant
   protocols ought to either change the value of the <x:ref>Content-Type</x:ref>
   header field or decode the representation before forwarding the message.
   (Some experimental applications of Content-Type for Internet mail have used
   a media-type parameter of ";conversions=&lt;content-coding&gt;" to perform
   a function equivalent to Content-Encoding. However, this parameter is
   not part of the MIME standards.)
</t>
</section>

<section title="Conversion of Content-Transfer-Encoding" anchor="conversion.of.content-transfer-encoding">
  <iref item="Content-Transfer-Encoding header field" x:for-anchor=""/>
<t>
   HTTP does not use the Content-Transfer-Encoding field of MIME.
   Proxies and gateways from MIME-compliant protocols to HTTP need to remove
   any Content-Transfer-Encoding prior to delivering the response message to
   an HTTP client.
</t>
<t>
   Proxies and gateways from HTTP to MIME-compliant protocols are
   responsible for ensuring that the message is in the correct format
   and encoding for safe transport on that protocol, where "safe
   transport" is defined by the limitations of the protocol being used.
   Such a proxy or gateway ought to transform and label the data with an
   appropriate Content-Transfer-Encoding if doing so will improve the
   likelihood of safe transport over the destination protocol.
</t>
</section>

<section title="MHTML and Line Length Limitations" anchor="mhtml.line.length">
<t>
   HTTP implementations that share code with MHTML <xref target="RFC2557"/>
   implementations need to be aware of MIME line length limitations.
   Since HTTP does not have this limitation, HTTP does not fold long lines.
   MHTML messages being transported by HTTP follow all conventions of MHTML,
   including line length limitations and folding, canonicalization, etc.,
   since HTTP transfers message-bodies without modification and, aside from the
   "multipart/byteranges" type (<xref target="multipart.byteranges"/>),
   does not interpret
   the content or any MIME header lines that might be contained therein.
</t>
</section>
</section>

<section title="Changes from Previous RFCs" anchor="changes">
<section title="Changes from HTTP/0.9" anchor="changes.from.0.9">
<t>
   Since HTTP/0.9 did not support header fields in a request, there is no
   mechanism for it to support name-based virtual hosts (selection of resource
   by inspection of the <x:ref>Host</x:ref> header field).
   Any server that implements name-based virtual hosts ought to disable
   support for HTTP/0.9. Most requests that appear to be HTTP/0.9 are, in
   fact, badly constructed HTTP/1.x requests caused by a client failing to
   properly encode the request-target.
</t>
</section>

<section title="Changes from HTTP/1.0" anchor="changes.from.1.0">
<section title="Multihomed Web Servers" anchor="changes.to.simplify.multihomed.web.servers.and.conserve.ip.addresses">
<t>
   The requirements that clients and servers support the <x:ref>Host</x:ref>
   header field (<xref target="field.host"/>), report an error if it is
   missing from an HTTP/1.1 request, and accept absolute URIs
   (<xref target="request.target"/>)
   are among the most important changes defined by HTTP/1.1.
</t>
<t>
   Older HTTP/1.0 clients assumed a one-to-one relationship of IP
   addresses and servers; there was no established mechanism for
   distinguishing the intended server of a request other than the IP address
   to which that request was directed. The <x:ref>Host</x:ref> header field was
   introduced during the development of HTTP/1.1 and, though it was
   quickly implemented by most HTTP/1.0 browsers, additional requirements
   were placed on all HTTP/1.1 requests in order to ensure complete
   adoption.  At the time of this writing, most HTTP-based services
   are dependent upon the Host header field for targeting requests.
</t>
</section>

<section title="Keep-Alive Connections" anchor="compatibility.with.http.1.0.persistent.connections">
<t>
   In HTTP/1.0, each connection is established by the client prior to the
   request and closed by the server after sending the response. However, some
   implementations implement the explicitly negotiated ("Keep-Alive") version
   of persistent connections described in <xref x:sec="19.7.1" x:fmt="of"
   target="RFC2068"/>.
</t>
<t>
   Some clients and servers might wish to be compatible with these previous
   approaches to persistent connections, by explicitly negotiating for them
   with a "Connection: keep-alive" request header field. However, some
   experimental implementations of HTTP/1.0 persistent connections are faulty;
   for example, if an HTTP/1.0 proxy server doesn't understand
   <x:ref>Connection</x:ref>, it will erroneously forward that header field
   to the next inbound server, which would result in a hung connection.
</t>
<t>
   One attempted solution was the introduction of a Proxy-Connection header
   field, targeted specifically at proxies. In practice, this was also
   unworkable, because proxies are often deployed in multiple layers, bringing
   about the same problem discussed above.
</t>
<t>
   As a result, clients are encouraged not to send the Proxy-Connection header
   field in any requests.
</t>
<t>
   Clients are also encouraged to consider the use of "Connection: keep-alive"
   in requests carefully; while they can enable persistent connections with
   HTTP/1.0 servers, clients using them will need to monitor the
   connection for "hung" requests (which indicate that the client ought to stop
   sending the header field), and this mechanism ought not be used by clients
   at all when a proxy is being used.
</t>
</section>

<section title="Introduction of Transfer-Encoding" anchor="introduction.of.transfer-encoding">
<t>
   HTTP/1.1 introduces the <x:ref>Transfer-Encoding</x:ref> header field
   (<xref target="field.transfer-encoding"/>).
   Transfer codings need to be decoded prior to forwarding an HTTP message
   over a MIME-compliant protocol.
</t>
</section>
</section>

<section title="Changes from RFC 7230" anchor="changes.from.rfc.7230">
<t>
   Most of the sections introducing HTTP's design goals, history, architecture,
   conformance criteria, protocol versioning, URIs, message routing, and
   header fields have been moved to <xref target="HTTP"/>.
   This document has been reduced to just the messaging syntax and
   connection management requirements specific to HTTP/1.1.
</t>
<t>
   Bare CRs have been prohibited outside of content.
   (<xref target="message.parsing"/>)
</t>
<t>
   The ABNF definition of <x:ref>authority-form</x:ref> has changed from the
   more general authority component of a URI (in which port is optional) to
   the specific host:port format that is required by CONNECT.
   (<xref target="authority-form"/>)
</t>
<t>
   Recipients are required to avoid smuggling/splitting attacks when processing an
   ambiguous message framing.
   (<xref target="field.transfer-encoding"/>)
</t>
<t>
   In the ABNF for chunked extensions, (bad) whitespace around ";" and "="
   has been reintroduced. Whitespace was removed
   in <xref target="RFC7230"/>, but that change was found to break existing
   implementations.
   (<xref target="chunked.extension"/>)
</t>
<t>
   Trailer field semantics now transcend the specifics of chunked transfer coding.
   The decoding algorithm for chunked (<xref target="decoding.chunked"/>) has
   been updated to encourage storage/forwarding of trailer fields separately
   from the header section, to only allow merging into the header section if
   the recipient knows the corresponding field definition permits and defines
   how to merge, and otherwise to discard the trailer fields instead of
   merging. The trailer part is now called the trailer section to be more
   consistent with the header section and more distinct from a body part.
   (<xref target="chunked.trailer.section"/>)
</t>
<t>
   Transfer coding parameters called "q" are disallowed in order to avoid
   conflicts with the use of ranks in the <x:ref>TE</x:ref> header field.
   (<xref target="transfer.coding.registry"/>)
</t>
</section>
</section>

<section title="Acknowledgements" anchor="acks" numbered="false">
<t>
  See <xref target="HTTP" x:rel="#acks"/>, which applies to this document as well.
</t>
</section>
</back>
</rfc>