<?xml version="1.0" encoding="UTF-8"?>
<!--
This XML document is the output of clean-for-DTD.xslt; a tool that strips
extensions to RFC 7749 from documents for processing with xml2rfc. draft submitted in xml v3 -->
<!--TARGET-GENERATOR: 202007-->
<!--TARGET-VOCABULARY: 3-->
<?xml-stylesheet type='text/xsl' href='lib/myxml2rfc.xslt'?>
<?rfc toc="yes" ?>
<?rfc symrefs="yes" ?>
<?rfc sortrefs="yes" ?>
<?rfc compact="yes"?>
<?rfc subcompact="no" ?>
<?rfc linkmailto="no" ?>
<?rfc editing="no" ?>
<?rfc comments="yes"?>
<?rfc inline="yes"?>
<?rfc rfcedstyle="yes"?>
<!DOCTYPE rfc [
<!ENTITY nbsp " ">
<!ENTITY zwsp "​">
<!ENTITY nbhy "‑">
<!ENTITY wj "⁠">
]>
<rfc version="3"
tocInclude="true"
tocDepth="4"
sortRefs="true"
symRefs="true"
submissionType="IETF"
category="std"
consensus="true"
ipr="pre5378Trust200902"
obsoletes="7230"
docName="draft-ietf-httpbis-messaging-19">
<!--see https://trac.tools.ietf.org/tools/xml2rfc/trac/ticket/420-->
<?v3xml2rfc silence="Warning: Setting consensus="true" for IETF STD document"?>
<?v3xml2rfc silence="Warning: Expected a valid submissionType (stream) setting"?>
updates=""
docName="draft-ietf-httpbis-messaging-19"
number="9112"
xmlns:xi="http://www.w3.org/2001/XInclude"
xml:lang="en">
<front>
<title>HTTP/1.1</title>
<seriesInfo name="RFC" value="9112"/>
<seriesInfo name="STD" value="99"/>
<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>
<postalLine>United States of America</postalLine>
</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 VIC</postalLine>
<postalLine>Prahran</postalLine>
<postalLine>Australia</postalLine>
</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 Münster</postalLine>
<postalLine>Germany</postalLine>
</postal>
<email>julian.reschke@greenbytes.de</email>
<uri>https://greenbytes.de/tech/webdav/</uri>
</address>
</author>
<date year="2021" month="September" day="10"/> year="2022" month="June"/>
<area>Applications and Real-Time</area>
<workgroup>HTTP Working Group</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>
<note title="Editorial Note">
<t>This note is to be removed before publishing as an RFC.</t>
<t>
Discussion of this draft takes place on the HTTP working group
mailing list (ietf-http-wg@w3.org), which is archived at
<eref target="https://lists.w3.org/Archives/Public/ietf-http-wg/"
brackets="angle"/>.
</t>
<t>
Working Group information can be found at <eref target="https://httpwg.org/" brackets="angle"/>;
source code and issues list for this draft can be found at
<eref target="https://github.com/httpwg/http-core" brackets="angle"/>.
</t>
<t>
The changes in this draft are summarized in <xref target="changes.since.18"/>.
</t>
</note>
</front>
<middle>
<section anchor="introduction" title="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 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 anchor="requirements.notation" title="Requirements Notation">
<t>
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL
NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED",
"MAY", "<bcp14>MUST</bcp14>", "<bcp14>MUST NOT</bcp14>", "<bcp14>REQUIRED</bcp14>", "<bcp14>SHALL</bcp14>", "<bcp14>SHALL
NOT</bcp14>", "<bcp14>SHOULD</bcp14>", "<bcp14>SHOULD NOT</bcp14>", "<bcp14>RECOMMENDED</bcp14>", "<bcp14>NOT RECOMMENDED</bcp14>",
"<bcp14>MAY</bcp14>", and "OPTIONAL" "<bcp14>OPTIONAL</bcp14>" in this document are to be interpreted as
described in BCP 14 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" section="2"/>.
</t>
</section>
<section anchor="notation" title="Syntax 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="HTTP" section="5.6.1"/>,
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"
obsolete grammar rules that appear for historical reasons.
</t>
<t anchor="core.rules">
The following core rules are included by
reference, as defined in <xref target="RFC5234" sectionFormat="comma" section="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">
The rules below are defined in <xref target="HTTP"/>:
</t>
<sourcecode type="abnf7230"><![CDATA[ type="abnf9110"><![CDATA[ BWS = <BWS, see [HTTP], Section 5.6.3>
OWS = <OWS, see [HTTP], Section 5.6.3>
RWS = <RWS, see [HTTP], Section 5.6.3>
absolute-path = <absolute-path, see [HTTP], Section 4.1>
field-name = <field-name, see [HTTP], Section 5.1>
field-value = <field-value, see [HTTP], Section 5.5>
obs-text = <obs-text, see [HTTP], Section 5.6.4>
quoted-string = <quoted-string, see [HTTP], Section 5.6.4>
token = <token, see [HTTP], Section 5.6.2>
transfer-coding =
<transfer-coding, see [HTTP], Section 10.1.4>
]]></sourcecode>
<t anchor="imported.uri.rules">
The rules below are defined in <xref target="URI"/>:
</t>
<sourcecode type="abnf7230"><![CDATA[ type="abnf9110"><![CDATA[ absolute-URI = <absolute-URI, see [URI], Section 4.3>
authority = <authority, see [URI], Section 3.2>
uri-host = <host, see [URI], Section 3.2.2>
port = <port, see [URI], Section 3.2.3>
query = <query, see [URI], Section 3.4>
]]></sourcecode>
</section>
</section>
<section anchor="http.message" title="Message">
<t>
HTTP/1.1 clients and servers communicate by sending messages.
See <xref target="HTTP" section="3"/> for
the general terminology and core concepts of HTTP.
</t>
<section anchor="message.format" title="Message Format">
<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>
<iref primary="true" item="Grammar" subitem="HTTP-message"/>
<sourcecode type="abnf7230"><![CDATA[ type="abnf9110"><![CDATA[ HTTP-message = start-line CRLF
*( field-line CRLF )
CRLF
[ message-body ]
]]></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 message 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>
<iref primary="true" item="Grammar" subitem="start-line"/>
<sourcecode type="abnf7230"><![CDATA[ type="abnf9110"><![CDATA[ start-line = request-line / status-line
]]></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) 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 anchor="message.parsing" title="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 <bcp14>MUST</bcp14> 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 <bcp14>MAY</bcp14> recognize a
single LF as a line terminator and ignore any preceding CR.
</t>
<t>
A sender <bcp14>MUST NOT</bcp14> 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 <bcp14>MUST</bcp14> 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 <bcp14>MUST NOT</bcp14>
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 <bcp14>MUST</bcp14> 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 <bcp14>SHOULD</bcp14> ignore at least one empty line (CRLF)
received prior to the request-line.
</t>
<t>
A sender <bcp14>MUST NOT</bcp14> 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 <bcp14>MUST</bcp14> 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 <bcp14>SHOULD</bcp14> respond with a 400 (Bad Request) response
and close the connection.
</t>
</section>
<section anchor="http.version" title="HTTP Version">
<t>
HTTP uses a "<major>.<minor>" numbering scheme to indicate
versions of the protocol. This specification defines version "1.1".
<xref target="HTTP" section="2.5"/> 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 <xref target="message.format" format="none">start-line</xref>. HTTP-version is case-sensitive.
</t>
<iref primary="true" item="Grammar" subitem="HTTP-version"/>
<iref primary="true" item="Grammar" subitem="HTTP-name"/>
<sourcecode type="abnf7230"><![CDATA[ type="abnf9110"><![CDATA[ HTTP-version = HTTP-name "/" DIGIT "." DIGIT
HTTP-name = %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) <bcp14>MUST</bcp14> 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 <xref target="message.format" format="none">start-line</xref> 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 <bcp14>MAY</bcp14> 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 <bcp14>SHOULD NOT</bcp14> be
performed unless triggered by specific client attributes, such as when
one or more of the request header fields (e.g., User-Agent)
uniquely match the values sent by a client known to be in error.
</t>
</section>
</section>
<section anchor="request.line" title="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>
<iref primary="true" item="Grammar" subitem="request-line"/>
<sourcecode type="abnf7230"><![CDATA[ type="abnf9110"><![CDATA[ request-line = method SP request-target SP HTTP-version
]]></sourcecode>
<t>
Although the request-line grammar rule requires that each of the component
elements be separated by a single SP octet, recipients <bcp14>MAY</bcp14> 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" section="2.3"/>.
A server that receives a method longer than any that it implements
<bcp14>SHOULD</bcp14> respond with a 501 (Not Implemented) status code.
A server that receives a request-target longer than any URI it wishes to
parse <bcp14>MUST</bcp14> respond with a
414 (URI Too Long) status code (see <xref target="HTTP" section="15.5.15"/>).
</t>
<t>
Various ad hoc limitations on request-line length are found in practice.
It is <bcp14>RECOMMENDED</bcp14> that all HTTP senders and recipients support, at a
minimum, request-line lengths of 8000 octets.
</t>
<section anchor="request.method" title="Method">
<iref primary="true" item="method"/>
<t>
The method token indicates the request method to be performed on the
target resource. The request method is case-sensitive.
</t>
<iref primary="true" item="Grammar" subitem="method"/>
<sourcecode type="abnf7230"><![CDATA[ type="abnf9110"><![CDATA[ method = token
]]></sourcecode>
<t>
The request methods defined by this specification can be found in
<xref target="HTTP" section="9"/>, along with information regarding the HTTP method
registry and considerations for defining new methods.
</t>
</section>
<section anchor="request.target" title="Request Target">
<iref primary="true" item="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>
<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"/>
<sourcecode type="abnf7230"><![CDATA[ type="abnf9110"><![CDATA[ request-target = origin-form
/ absolute-form
/ authority-form
/ asterisk-form
]]></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 <bcp14>SHOULD</bcp14> respond with either a
400 (Bad Request) error or a 301 (Moved Permanently)
redirect with the request-target properly encoded. A recipient <bcp14>SHOULD NOT</bcp14>
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 <bcp14>MUST</bcp14> send a Host header field (<xref target="HTTP" section="7.2"/>)
in all HTTP/1.1 request messages. If the target URI includes an authority component, then a client <bcp14>MUST</bcp14>
send a field value for Host that is identical to that authority
component, excluding any userinfo subcomponent and its "@" delimiter
(<xref target="HTTP" section="4.2.1"/>). section="4.2"/>).
If the authority component is missing or undefined for the target URI,
then a client <bcp14>MUST</bcp14> send a Host header field with an empty field value.
</t>
<t>
A server <bcp14>MUST</bcp14> respond with a 400 (Bad Request) 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 anchor="origin-form" title="origin-form">
<iref item="origin-form (of request-target)"/>
<t>
The most common form of request-target is the <em>origin-form</em>. "origin-form".
</t>
<iref primary="true" item="Grammar" subitem="origin-form"/>
<sourcecode type="abnf7230"><![CDATA[ type="abnf9110"><![CDATA[ origin-form = absolute-path [ "?" query ]
]]></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 <bcp14>MUST</bcp14> 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 <bcp14>MUST</bcp14> send
"/" as the path within the origin-form of request-target.
A Host header field is also sent, as defined in
<xref target="HTTP" section="7.2"/>.
</t>
<t>
For example, a client wishing to retrieve a representation of the resource
identified as
</t>
<artwork type="example"><![CDATA[
<artwork><![CDATA[
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"><![CDATA[GET /where?q=now HTTP/1.1
Host: www.example.org
]]></sourcecode>
<t>
followed by the remainder of the request message.
</t>
</section>
<section anchor="absolute-form" title="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 <bcp14>MUST</bcp14> send the target URI
in <em>absolute-form</em> "absolute-form" as the request-target.
</t>
<iref primary="true" item="Grammar" subitem="absolute-form"/>
<sourcecode type="abnf7230"><![CDATA[ type="abnf9110"><![CDATA[ absolute-form = absolute-URI
]]></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 to 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="HTTP" section="7.6"/>.
</t>
<t>
An example absolute-form of request-line would be:
</t>
<sourcecode type="http-message"><![CDATA[GET http://www.example.org/pub/WWW/TheProject.html HTTP/1.1
]]></sourcecode>
<t>
A client <bcp14>MUST</bcp14> 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 <bcp14>MUST</bcp14> 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
<bcp14>MUST</bcp14> 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 <bcp14>MUST</bcp14> 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 <bcp14>MUST</bcp14> 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 anchor="authority-form" title="authority-form">
<iref item="authority-form (of request-target)"/>
<t>
The <em>authority-form</em> "authority-form" of request-target is only used for
CONNECT requests (<xref target="HTTP" section="9.3.6"/>). It consists of only the
uri-host and port number of the tunnel
destination, separated by a colon (":").
</t>
<iref primary="true" item="Grammar" subitem="authority-form"/>
<sourcecode type="abnf7230"><![CDATA[ type="abnf9110"><![CDATA[ authority-form = uri-host ":" port
]]></sourcecode>
<t>
When making a CONNECT request to establish a tunnel through one or more
proxies, a client <bcp14>MUST</bcp14> 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 <xref target="imported.uri.rules" format="none">authority</xref> 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"><![CDATA[CONNECT www.example.com:80 HTTP/1.1
Host: www.example.com
]]></sourcecode>
</section>
<section anchor="asterisk-form" title="asterisk-form">
<iref item="asterisk-form (of request-target)"/>
<t>
The <em>asterisk-form</em> "asterisk-form" of request-target is only used for a server-wide
OPTIONS request (<xref target="HTTP" section="9.3.7"/>).
</t>
<iref primary="true" item="Grammar" subitem="asterisk-form"/>
<sourcecode type="abnf7230"><![CDATA[ type="abnf9110"><![CDATA[ asterisk-form = "*"
]]></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 <bcp14>MUST</bcp14> send only "*" (%x2A) as the request-target.
For example,
</t>
<sourcecode type="http-message"><![CDATA[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 <bcp14>MUST</bcp14> 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"><![CDATA[OPTIONS http://www.example.org:8001 HTTP/1.1
]]></sourcecode>
<t>
would be forwarded by the final proxy as
</t>
<sourcecode type="http-message"><![CDATA[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 anchor="reconstructing.target.uri"
title="Reconstructing the Target URI">
<t>
The target URI is the <xref target="request.target" format="none">request-target</xref> when the
request-target is in <xref target="absolute-form" format="none">absolute-form</xref>. 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="HTTP" section="7.1"/>):
</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 <xref target="authority-form" format="none">authority-form</xref>,
the target URI's authority component is the request-target.
Otherwise, the target URI's authority component is the field value of the
Host header field. If there is no Host
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 <xref target="authority-form" format="none">authority-form</xref> or
<xref target="asterisk-form" format="none">asterisk-form</xref>, the target URI's combined
path and <xref target="imported.uri.rules" format="none">query</xref> component is empty.
Otherwise, the target URI's combined path and
<xref target="imported.uri.rules" format="none">query</xref> component is the request-target.
</li>
<li>
The components of a reconstructed target URI, once determined as above,
can be recombined into <xref target="imported.uri.rules" format="none">absolute-URI</xref> form by concatenating
the scheme, "://", authority, and combined path and query component.
</li>
</ul>
<t>
Example 1: the The following message received over a secure connection
</t>
<sourcecode type="http-message"><![CDATA[GET /pub/WWW/TheProject.html HTTP/1.1
Host: www.example.org
]]></sourcecode>
<t>
has a target URI of
</t>
<artwork type="example"><![CDATA[
<artwork><![CDATA[
https://www.example.org/pub/WWW/TheProject.html
]]></artwork>
<t>
Example 2: the The following message received over an insecure connection
</t>
<sourcecode type="http-message"><![CDATA[OPTIONS * HTTP/1.1
Host: www.example.org:8080
]]></sourcecode>
<t>
has a target URI of
</t>
<artwork type="example"><![CDATA[
<artwork><![CDATA[
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 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 <bcp14>MAY</bcp14> 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="HTTP" section="7.4"/>.
</t>
</section>
</section>
<section anchor="status.line" title="Status Line">
<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, space
and ending with an <bcp14>OPTIONAL</bcp14> textual phrase describing the status code.
</t>
<iref primary="true" item="Grammar" subitem="status-line"/>
<sourcecode type="abnf7230"><![CDATA[ type="abnf9110"><![CDATA[ status-line = HTTP-version SP status-code SP [reason-phrase] [ reason-phrase ]
]]></sourcecode>
<t>
Although the status-line grammar rule requires that each of the component
elements be separated by a single SP octet, recipients <bcp14>MAY</bcp14> 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>
<iref primary="true" item="Grammar" subitem="status-code"/>
<sourcecode type="abnf7230"><![CDATA[ type="abnf9110"><![CDATA[ status-code = 3DIGIT
]]></sourcecode>
<t>
HTTP's core status codes are defined in <xref target="HTTP" section="15"/>,
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>
<iref primary="true" item="Grammar" subitem="reason-phrase"/>
<sourcecode type="abnf7230"><![CDATA[ type="abnf9110"><![CDATA[ reason-phrase = 1*( HTAB / SP / VCHAR / obs-text )
]]></sourcecode>
<t>
A client <bcp14>SHOULD</bcp14> 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 <bcp14>MUST</bcp14> 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 three octets SP CR LF). space).
</t>
</section>
<section anchor="header.field.syntax" title="Field Syntax">
<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>
<iref primary="true" item="Grammar" subitem="field-line"/>
<iref primary="false" item="Grammar" subitem="field-name"/>
<iref primary="false" item="Grammar" subitem="field-value"/>
<sourcecode type="abnf7230"><![CDATA[ type="abnf9110"><![CDATA[ field-line = field-name ":" OWS field-value OWS
]]></sourcecode>
<t>
Most HTTP field names and the rules
Rules for parsing within field values are
defined in <xref target="HTTP" section="6.3"/>. section="5.5"/>. This section covers the
generic syntax for header field inclusion within, and extraction from,
HTTP/1.1 messages.
</t>
<section anchor="field.parsing" title="Field Line 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 <bcp14>MUST</bcp14> reject, with a response status code of
400 (Bad Request), any received request message that contains
whitespace between a header field name and colon.
A proxy <bcp14>MUST</bcp14> 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 anchor="line.folding" title="Obsolete Line Folding">
<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 "message/http" media type
(<xref target="media.type.message.http"/>).
</t>
<iref primary="true" item="Grammar" subitem="obs-fold"/>
<sourcecode type="abnf7230"><![CDATA[ type="abnf9110"><![CDATA[ obs-fold = OWS CRLF RWS
; obsolete line folding
]]></sourcecode>
<t>
A sender <bcp14>MUST NOT</bcp14> generate a message that includes line folding
(i.e., that has any field line value that contains a match to the
<xref target="line.folding" format="none">obs-fold</xref> rule) unless the message is intended for packaging
within the message/http "message/http" media type.
</t>
<t>
A server that receives an <xref target="line.folding" format="none">obs-fold</xref> in a request message that
is not within a message/http "message/http" container <bcp14>MUST</bcp14> either reject the message by
sending a 400 (Bad Request), preferably with a
representation explaining that obsolete line folding is unacceptable, or
replace each received <xref target="line.folding" format="none">obs-fold</xref> with one or more
<xref target="core.rules" format="none">SP</xref> octets prior to interpreting the field value or
forwarding the message downstream.
</t>
<t>
A proxy or gateway that receives an <xref target="line.folding" format="none">obs-fold</xref> in a response
message that is not within a message/http "message/http" container <bcp14>MUST</bcp14> either discard
the message and replace it with a 502 (Bad Gateway)
response, preferably with a representation explaining that unacceptable
line folding was received, or replace each received <xref target="line.folding" format="none">obs-fold</xref>
with one or more <xref target="core.rules" format="none">SP</xref> octets prior to interpreting the field
value or forwarding the message downstream.
</t>
<t>
A user agent that receives an <xref target="line.folding" format="none">obs-fold</xref> in a response message
that is not within a message/http "message/http" container <bcp14>MUST</bcp14> replace each received
<xref target="line.folding" format="none">obs-fold</xref> with one or more <xref target="core.rules" format="none">SP</xref> octets prior to
interpreting the field value.
</t>
</section>
</section>
<section anchor="message.body" title="Message Body">
<t>
The message body (if any) of an HTTP/1.1 message is used to carry content
(<xref target="HTTP" section="6.4"/>) 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>
<iref primary="true" item="Grammar" subitem="message-body"/>
<sourcecode type="abnf7230"><![CDATA[ type="abnf9110"><![CDATA[ message-body = *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
<xref target="body.content-length" format="none">Content-Length</xref> or <xref target="field.transfer-encoding" format="none">Transfer-Encoding</xref> header
field. Request message framing is independent of method semantics.
</t>
<t>
The presence of a message body in a response 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 (<xref target="status.line"/>), and code.
This corresponds to when response content is
allowed; see <xref
allowed by HTTP semantics (<xref target="HTTP" section="6.4"/>. section="6.4.1"/>).
</t>
<section anchor="field.transfer-encoding" title="Transfer-Encoding">
<iref primary="true" item="Fields" subitem="Transfer-Encoding"/>
<iref primary="true" item="Header Fields" subitem="Transfer-Encoding"/>
<iref primary="true" item="Transfer-Encoding header field"/>
<iref item="chunked (Coding Format)"/>
<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>
<iref primary="true" item="Grammar" subitem="Transfer-Encoding"/>
<sourcecode type="abnf7230"><![CDATA[ type="abnf9110"><![CDATA[ Transfer-Encoding = #transfer-coding
; defined in [HTTP], Section 10.1.4
]]></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" sectionFormat="comma" section="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 <bcp14>MUST</bcp14> 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 <bcp14>MUST NOT</bcp14> 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 <bcp14>MUST</bcp14> 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 <bcp14>MUST</bcp14> 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"><![CDATA[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 Content-Encoding (<xref target="HTTP" section="8.4.1"/>),
Transfer-Encoding is a property of the message, not of the representation, and
any representation.
Any recipient along the request/response chain <bcp14>MAY</bcp14> 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 <bcp14>MAY</bcp14> be sent in a response to a HEAD request or in a
304 (Not Modified) response (<xref target="HTTP" section="15.4.5"/>) 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 <bcp14>MUST NOT</bcp14> send a Transfer-Encoding header field in any response
with a status code of
1xx (Informational) or 204 (No Content).
A server <bcp14>MUST NOT</bcp14> send a Transfer-Encoding header field in any
2xx (Successful) response to a CONNECT request (<xref target="HTTP" section="9.3.6"/>).
</t>
<t>
A server that receives a request message with a transfer coding it does
not understand <bcp14>SHOULD</bcp14> respond with 501 (Not Implemented).
</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 encoding transfer coding in transit.
</t>
<t>
A client <bcp14>MUST NOT</bcp14> 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 <bcp14>MUST NOT</bcp14> 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 encoding 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 <bcp14>MAY</bcp14> 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 <bcp14>MUST</bcp14> 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 <bcp14>MUST</bcp14> 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 anchor="body.content-length" title="Content-Length">
<iref primary="false" item="Content-Length header field"/>
<t>
When a message does not have a <xref target="field.transfer-encoding" format="none">Transfer-Encoding</xref> header
field, a Content-Length header field (<xref target="HTTP" section="8.6"/>) 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="HTTP" section="8.6"/>).
</t>
<t>
A sender <bcp14>MUST NOT</bcp14> send a Content-Length header field in any message that
contains a <xref target="field.transfer-encoding" format="none">Transfer-Encoding</xref> header field.
</t>
<aside>
<t>
<strong>Note:</strong> 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 anchor="message.body.length" title="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
1xx (Informational), 204 (No Content), or
304 (Not Modified) 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 2xx (Successful) 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 <bcp14>MUST</bcp14> ignore any
<xref target="body.content-length" format="none">Content-Length</xref> or <xref target="field.transfer-encoding" format="none">Transfer-Encoding</xref> header
fields received in such a message.
</t>
</li>
<li>
<t>
If a message is received with both a <xref target="field.transfer-encoding" format="none">Transfer-Encoding</xref>
and a <xref target="body.content-length" format="none">Content-Length</xref> 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 <bcp14>MUST</bcp14> 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 <xref target="field.transfer-encoding" format="none">Transfer-Encoding</xref> 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 <xref target="field.transfer-encoding" format="none">Transfer-Encoding</xref> 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 <xref target="field.transfer-encoding" format="none">Transfer-Encoding</xref> 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 <bcp14>MUST</bcp14> respond with
the 400 (Bad Request) status code and then close the
connection.
</t>
</li>
<li>
<t>
If a message is received without <xref target="field.transfer-encoding" format="none">Transfer-Encoding</xref> and with
an invalid <xref target="body.content-length" format="none">Content-Length</xref> header field, then the message
framing is invalid and the recipient <bcp14>MUST</bcp14> treat it as an unrecoverable
error, unless the field value can be successfully parsed as a
comma-separated list (<xref target="HTTP" section="5.6.1"/>), all values in the
list are valid, and all values in the list are the same (in which case 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 <bcp14>MUST</bcp14> respond with
a 400 (Bad Request) status code and then close the connection.
If it is in a response message received by a proxy,
the proxy <bcp14>MUST</bcp14> close the connection to the server, discard the received
response, and send a 502 (Bad Gateway) response to the
client.
If it is in a response message received by a user agent,
the user agent <bcp14>MUST</bcp14> close the connection to the server and discard the
received response.
</t>
</li>
<li>
<t>
If a valid <xref target="body.content-length" format="none">Content-Length</xref> header field is present without
<xref target="field.transfer-encoding" format="none">Transfer-Encoding</xref>, 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 <bcp14>MUST</bcp14> 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 <bcp14>SHOULD</bcp14> generate encoding or
length-delimited messages whenever possible. The close-delimiting
feature exists primarily for backwards compatibility with HTTP/1.0.
</t>
<aside>
<t>
<strong>Note:</strong> 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 <bcp14>MAY</bcp14> reject a request that contains a message body but
not a <xref target="body.content-length" format="none">Content-Length</xref> by responding with
411 (Length Required).
</t>
<t>
Unless a transfer coding other than chunked has been applied,
a client that sends a request containing a message body <bcp14>SHOULD</bcp14>
use a valid <xref target="body.content-length" format="none">Content-Length</xref> 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 411 (Length Required)
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 content length in advance of being called 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 <bcp14>MUST</bcp14> send
either a valid <xref target="body.content-length" format="none">Content-Length</xref> header field or use the
chunked transfer coding. A client <bcp14>MUST NOT</bcp14> use the chunked transfer
encoding
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
<bcp14>MAY</bcp14> 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 <bcp14>MUST NOT</bcp14>
process, cache, or forward such extra data as a separate response, since
such behavior would be vulnerable to cache poisoning.
</t>
</section>
</section>
<section anchor="transfer.codings" title="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 <xref target="field.transfer-encoding" format="none">Transfer-Encoding</xref>
(<xref target="field.transfer-encoding"/>) and TE
(<xref target="HTTP" section="10.1.4"/>) header fields (the latter also
defining the "transfer-coding" grammar).
</t>
<section anchor="chunked.encoding" title="Chunked Transfer Coding">
<iref primary="true" item="chunked (transfer coding)"/>
<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
<bcp14>OPTIONAL</bcp14> 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>
<iref primary="true" item="Grammar" subitem="chunked-body"/>
<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"/>
<sourcecode type="abnf7230"><![CDATA[ type="abnf9110"><![CDATA[ chunked-body = *chunk
last-chunk
trailer-section
CRLF
chunk = chunk-size [ chunk-ext ] CRLF
chunk-data CRLF
chunk-size = 1*HEXDIG
last-chunk = 1*("0") [ chunk-ext ] CRLF
chunk-data = 1*OCTET ; 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 <bcp14>MUST</bcp14> 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 <bcp14>MUST</bcp14> 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 encoding coding does not define any parameters. Their presence
<bcp14>SHOULD</bcp14> be treated as an error.
</t>
<section anchor="chunked.extension" title="Chunk Extensions">
<t>
The chunked encoding coding allows each chunk to include zero or more chunk
extensions, immediately following the <xref target="chunked.encoding" format="none">chunk-size</xref>, 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>
<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"/>
<sourcecode type="abnf7230"><![CDATA[ type="abnf9110"><![CDATA[ chunk-ext = *( BWS ";" BWS chunk-ext-name
[ BWS "=" BWS chunk-ext-val ] )
chunk-ext-name = token
chunk-ext-val = token / quoted-string
]]></sourcecode>
<t>
The chunked encoding 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 <bcp14>MUST</bcp14> 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 4xx (Client Error)
response if that amount is exceeded.
</t>
</section>
<section anchor="chunked.trailer.section" title="Chunked 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="HTTP" section="6.5"/>.
</t>
<iref primary="true" item="Grammar" subitem="trailer-section"/>
<iref primary="false" item="Grammar" subitem="field-line"/>
<sourcecode type="abnf7230"><![CDATA[ type="abnf9110"><![CDATA[ trailer-section = *( field-line CRLF )
]]></sourcecode>
<t>
A recipient that removes the chunked encoding coding from a message <bcp14>MAY</bcp14>
selectively retain or discard the received trailer fields. A recipient
that retains a received trailer field <bcp14>MUST</bcp14> either store/forward the
trailer field separately from the received header fields or merge the
received trailer field into the header section.
A recipient <bcp14>MUST NOT</bcp14> 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 anchor="decoding.chunked" title="Decoding Chunked">
<t>
A process for decoding the chunked transfer coding
can be represented in pseudo-code as:
</t>
<sourcecode type="pseudocode"><![CDATA[ length := 0
read chunk-size, chunk-ext (if any), and CRLF
while (chunk-size > 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 anchor="compression.codings" title="Transfer Codings for Compression">
<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="HTTP" section="8.4.1.1"/>.</dd>
<dt>deflate</dt>
<dd>See <xref target="HTTP" section="8.4.1.2"/>.</dd>
<dt>gzip (and x-gzip)</dt>
<dd>See <xref target="HTTP" section="8.4.1.3"/>.</dd>
</dl>
<t>
The compression codings do not define any parameters. The presence
of parameters with any of these compression codings <bcp14>SHOULD</bcp14> be treated
as an error.
</t>
</section>
<section anchor="transfer.coding.registry" title="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"
brackets="angle"/>.
</t>
<t>
Registrations <bcp14>MUST</bcp14> include the following fields:
</t>
<ul>
<li>Name</li>
<li>Description</li>
<li>Pointer to specification text</li>
</ul>
<t>
Names of transfer codings <bcp14>MUST NOT</bcp14> overlap with names of content codings
(<xref target="HTTP" section="8.4.1"/>) unless the encoding transformation is identical, as
is the case for the compression codings defined in
<xref target="compression.codings"/>.
</t>
<t>
The TE header field (<xref target="HTTP" section="10.1.4"/>) uses a
pseudo parameter
pseudo-parameter named "q" as the rank value when multiple transfer codings
are acceptable. Future registrations of transfer codings <bcp14>SHOULD NOT</bcp14>
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" section="4.8"/>), section="4.8"/>) and <bcp14>MUST</bcp14>
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 anchor="transfer.coding.negotiation"
title="Negotiating Transfer Codings">
<t>
The TE field (<xref target="HTTP" section="10.1.4"/>) is used in HTTP/1.1 to indicate
what transfer-codings, transfer codings, besides chunked, the client is willing to accept
in the response, response and whether the client is willing to preserve
trailer fields in a chunked transfer coding.
</t>
<t>
A client <bcp14>MUST NOT</bcp14> 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"><![CDATA[TE: deflate
TE:
TE: trailers, deflate;q=0.5
]]></sourcecode>
<t>
When multiple transfer codings are acceptable, the client <bcp14>MAY</bcp14> rank the
codings by preference using a case-insensitive "q" parameter (similar to
the qvalues used in content negotiation fields, fields; see <xref target="HTTP" section="12.4.2"/>). 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="HTTP" section="6.5"/>.
</t>
<t>
Since the TE header field only applies to the immediate connection,
a sender of TE <bcp14>MUST</bcp14> also send a "TE" connection option within the
Connection header field (<xref target="HTTP" section="7.6.1"/>)
in order to prevent the TE header field from being forwarded by intermediaries
that do not support its semantics.
</t>
</section>
</section>
<section anchor="incomplete.messages" title="Handling Incomplete Messages">
<t>
A server that receives an incomplete request message, usually due to a
canceled request or a triggered timeout exception, <bcp14>MAY</bcp14> 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, <bcp14>MUST</bcp14> record the message as incomplete.
Cache requirements for incomplete responses are defined in
<xref target="CACHING" section="3"/>. section="3.3"/>.
</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 <xref target="body.content-length" format="none">Content-Length</xref> 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 anchor="connection.management" title="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="HTTP" section="7.3"/>, the specific
connection protocols to be used for an HTTP interaction are determined by
client configuration and the target URI.
For example, the "http" URI scheme
(<xref target="HTTP" section="4.2.1"/>) 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 anchor="persistent.establishment" title="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 anchor="associating.response.to.request"
title="Associating a Response to a 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 (1xx, (1xx; see <xref target="HTTP" section="15.2"/>) precede a
final response to the same request.
</t>
<t>
A client that has more than one outstanding request on a connection <bcp14>MUST</bcp14>
maintain a list of outstanding requests in the order sent and <bcp14>MUST</bcp14>
associate each received response message on that connection to the
first outstanding request that has not yet received a final
(non-1xx) response.
</t>
<t>
If a client receives data on a connection that doesn't have
outstanding requests, the client <bcp14>MUST NOT</bcp14> consider that data to be a
valid response; the client <bcp14>SHOULD</bcp14> close the connection, since message
delimitation is now ambiguous, unless the data consists only of one or
more CRLF (which can be discarded, as discarded per
<xref target="message.parsing"/>).
</t>
</section>
<section anchor="persistent.connections" title="Persistence">
<iref primary="false" item="close"/>
<t>
HTTP/1.1 defaults to the use of <em>persistent connections</em>, "persistent connections",
allowing multiple requests and responses to be carried over a single
connection. HTTP implementations <bcp14>SHOULD</bcp14> support persistent connections.
</t>
<t>
A recipient determines whether a connection is persistent or not based on
the protocol version and Connection header field
(<xref target="HTTP" section="7.6.1"/>) in the
most recently received message, if any:
</t>
<ul>
<li>If the <xref "<xref target="persistent.tear-down" format="none">close</xref> format="none">close</xref>" 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 <xref target="persistent.connections" format="none">persistent connections</xref>
<bcp14>MUST</bcp14>
send the <xref "<xref target="persistent.tear-down" format="none">close</xref> format="none">close</xref>" connection option in every request message.
</t>
<t>
A server that does not support <xref target="persistent.connections" format="none">persistent connections</xref>
<bcp14>MUST</bcp14>
send the <xref "<xref target="persistent.tear-down" format="none">close</xref> format="none">close</xref>" connection option in every response message
that does not have a 1xx (Informational) status code.
</t>
<t>
A client <bcp14>MAY</bcp14> send additional requests on a persistent connection until it
sends or receives a <xref "<xref target="persistent.tear-down" format="none">close</xref> format="none">close</xref>" 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 <bcp14>MUST</bcp14> read the entire request message body or close
the connection after sending its response, since otherwise response; otherwise, the
remaining data on a persistent connection would be misinterpreted
as the next request. Likewise,
a client <bcp14>MUST</bcp14> read the entire response message body if it intends
to reuse the same connection for a subsequent request.
</t>
<t>
A proxy server <bcp14>MUST NOT</bcp14> maintain a persistent connection with an
HTTP/1.0 client (see <xref target="RFC2068" section="19.7.1"/> 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 anchor="persistent.retrying.requests" title="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="HTTP" section="9.2.2"/>.
</t>
</section>
<section anchor="pipelining" title="Pipelining">
<t>
A client that supports persistent connections <bcp14>MAY</bcp14>
<em>pipeline</em> "pipeline"
its requests (i.e., send multiple requests without waiting for each
response). A server <bcp14>MAY</bcp14> process a sequence of pipelined requests in
parallel if they all have safe methods (<xref target="HTTP" section="9.2.1"/>), but it <bcp14>MUST</bcp14> send
the corresponding responses in the same order that the requests were
received.
</t>
<t>
A client that pipelines requests <bcp14>SHOULD</bcp14> 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 <bcp14>MUST NOT</bcp14> 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="HTTP" section="9.2.2"/>) are significant to pipelining
because they can be automatically retried after a connection failure.
A user agent <bcp14>SHOULD NOT</bcp14> 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 <bcp14>MAY</bcp14> 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 <bcp14>MAY</bcp14> 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 <bcp14>SHOULD</bcp14> 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 anchor="persistent.concurrency" title="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 anchor="persistent.failures" title="Failures and Timeouts">
<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 <bcp14>SHOULD</bcp14> issue a graceful close
on the connection. Implementations <bcp14>SHOULD</bcp14> 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 <bcp14>MAY</bcp14> 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 <bcp14>SHOULD</bcp14> sustain persistent connections, when possible, and allow
the underlying transport's flow-control mechanisms to resolve temporary overloads, 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 <bcp14>SHOULD</bcp14> 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 <bcp14>SHOULD</bcp14> immediately cease transmitting the body and close its side
of the connection.
</t>
</section>
<section anchor="persistent.tear-down" title="Tear-down">
<iref primary="false" item="Connection header field"/>
<iref primary="true" item="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 <bcp14>SHOULD</bcp14> send a Connection header field
(<xref target="HTTP" section="7.6.1"/>) containing the close "close" connection option
when it intends to close a connection. For example,
</t>
<sourcecode type="http-message"><![CDATA[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 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"/>
Note that the field name "Close" is reserved, since using that name as a
header field might conflict with the close "close" connection option.
</t>
<t>
A client that sends a close "close" connection option <bcp14>MUST NOT</bcp14>
send further requests on that connection (after the one containing the
close)
"close") and <bcp14>MUST</bcp14> close the connection after reading the
final response message corresponding to this request.
</t>
<t>
A server that receives a close "close" connection option <bcp14>MUST</bcp14>
initiate closure of the connection (see below) after it sends the
final response to the request that contained the close "close" connection option.
The server <bcp14>SHOULD</bcp14> send a close "close" connection option in its final response
on that connection. The server <bcp14>MUST NOT</bcp14> process any further requests
received on that connection.
</t>
<t>
A server that sends a close "close" connection option <bcp14>MUST</bcp14>
initiate closure of the connection (see below) after it sends the
response containing the close "close" connection option. The server <bcp14>MUST NOT</bcp14> process
any further requests received on that connection.
</t>
<t>
A client that receives a close "close" connection option <bcp14>MUST</bcp14>
cease sending requests on that connection and close the connection
after reading the response message containing the close "close" connection option;
if additional pipelined requests had been sent on the connection,
the client <bcp14>SHOULD NOT</bcp14> 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 anchor="tls.connection.initiation" title="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 <bcp14>MUST</bcp14> be sent as TLS
"application data", data" but is otherwise treated like a normal connection for
HTTP (including potential reuse as a persistent connection).
</t>
</section>
<section anchor="tls.connection.closure" title="TLS Connection Closure">
<t>
TLS uses an exchange of closure alerts prior to (non-error) connection
closure to provide secure connection closure; see <xref target="TLS13" section="6.1"/>. 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 were are
complete. Handling of incomplete messages is defined in
<xref target="incomplete.messages"/>.
</t>
<t>
When encountering an incomplete close, a client <bcp14>SHOULD</bcp14> treat as completed
all requests for which it has received either
</t>
<ol>
<li>
as much data as specified in the <xref target="body.content-length" format="none">Content-Length</xref> header or, when a
field or
</li>
<li>
the terminal zero-length chunk (when <xref target="field.transfer-encoding" format="none">Transfer-Encoding</xref> of chunked is used, for which the terminal
zero-length chunk has been received. 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 <bcp14>SHOULD</bcp14> recover gracefully.
</t>
<t>
Clients <bcp14>MUST</bcp14> send a closure alert before closing the connection.
Clients that do not expect to receive any more data <bcp14>MAY</bcp14> 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 <bcp14>SHOULD</bcp14> be prepared to receive an incomplete close from the client,
since the client can often determine when locate the end of server data is. data.
</t>
<t>
Servers <bcp14>MUST</bcp14> attempt to initiate an exchange of closure alerts with
the client before closing the connection. Servers <bcp14>MAY</bcp14> close the
connection after sending the closure alert, thus generating an
incomplete close on the client side.
</t>
</section>
</section>
<section anchor="enclosing.messages" title="Enclosing Messages as Data">
<section anchor="media.type.message.http" title="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 "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 "message/http" are allowed to use
line folding (<xref target="line.folding" format="none">obs-fold</xref>), as described in
<xref target="line.folding"/>, to convey the field value over multiple
lines. A recipient of message/http "message/http" data <bcp14>MUST</bcp14> 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 — -- "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>This specification
<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
<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 anchor="media.type.application.http"
title="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 "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 — -- "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>
This specification
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
<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 anchor="security.considerations" title="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 anchor="response.splitting" title="Response Splitting">
<t>
Response splitting (a.k.a., CRLF (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 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 Location 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, 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 anchor="request.smuggling" title="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 anchor="message.integrity" title="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 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 anchor="message.confidentiality" title="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="HTTP" section="4.2.2"/>.
</t>
</section>
</section>
<section anchor="IANA.considerations" title="IANA Considerations">
<t>
The change controller for the following registrations is:
"IETF (iesg@ietf.org) - Internet Engineering Task Force".
</t>
<section anchor="field.name.registration" title="Field Name Registration">
<t>
First, introduce
IANA has added the following field names to the new "Hypertext Transfer Protocol (HTTP) Field
Name Registry" at <eref target="https://www.iana.org/assignments/http-fields" brackets="angle"/> brackets="angle"/>,
as described in <xref target="HTTP" section="18.4"/>.
</t>
<t>
Then, please update the registry with the field names listed in the table
below:
</t>
<!--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>Ref.</th>
<th>Section</th>
<th>Comments</th>
</tr>
</thead>
<tbody>
<tr>
<td>Close</td>
<td>standard</td>
<td>permanent</td>
<td>
<xref target="persistent.tear-down" format="counter"/>
</td>
<td>(reserved)</td>
</tr>
<tr>
<td>MIME-Version</td>
<td>standard</td>
<td>permanent</td>
<td>
<xref target="mime-version" format="counter"/>
</td>
<td/>
</tr>
<tr>
<td>Transfer-Encoding</td>
<td>standard</td>
<td>permanent</td>
<td>
<xref target="field.transfer-encoding" format="counter"/>
</td>
<td/>
</tr>
</tbody>
</table>
<!--(END)-->
</section>
<section anchor="media.type.http" title="Media Type Registration">
<t>
Please update
IANA has updated the "Media Types" registry at
<eref target="https://www.iana.org/assignments/media-types" brackets="angle"/>
with the registration information in Sections
<xref target="media.type.message.http"/> target="media.type.message.http" format="counter"/> and
<xref target="media.type.application.http"/> target="media.type.application.http" format="counter"/> for the media types
"message/http" and "application/http", respectively.
</t>
</section>
<section anchor="transfer.coding.registration"
title="Transfer Coding Registration">
<t>
Please update
IANA has updated the "HTTP Transfer Coding Registry" at
<eref target="https://www.iana.org/assignments/http-parameters/"
brackets="angle"/>
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>Reference</th>
<th>Section</th>
</tr>
</thead>
<tbody>
<tr>
<td>chunked</td>
<td>Transfer in a series of chunks</td>
<td>
<xref target="chunked.encoding"/> 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"/> 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> target="RFC1950"/>)
</td>
<td>
<xref target="compression.codings"/> target="compression.codings" format="counter"/>
</td>
</tr>
<tr>
<td>gzip</td>
<td>GZIP file format <xref target="RFC1952"/>
</td>
<td>
<xref target="compression.codings"/> target="compression.codings" format="counter"/>
</td>
</tr>
<tr>
<td>trailers</td>
<td>(reserved)</td>
<td>
<xref target="transfer.coding.registration"/> target="transfer.coding.registration" format="counter"/>
</td>
</tr>
<tr>
<td>x-compress</td>
<td>Deprecated (alias for compress)</td>
<td>
<xref target="compression.codings"/> target="compression.codings" format="counter"/>
</td>
</tr>
<tr>
<td>x-gzip</td>
<td>Deprecated (alias for gzip)</td>
<td>
<xref target="compression.codings"/> target="compression.codings" format="counter"/>
</td>
</tr>
</tbody>
</table>
<aside>
<t>
<strong>Note:</strong> the coding name "trailers" is reserved because its use would
conflict with the keyword "trailers" in the TE
header field (<xref target="HTTP" section="10.1.4"/>).
</t>
</aside>
</section>
<section anchor="alpn.registration" title="ALPN Protocol ID Registration">
<t>
Please update
IANA has updated the
"TLS Application-Layer Protocol Negotiation (ALPN) Protocol IDs" registry at
<eref target="https://www.iana.org/assignments/tls-extensiontype-values/tls-extensiontype-values.xhtml" target="https://www.iana.org/assignments/tls-extensiontype-values/"
brackets="angle"/>
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>(this specification)</td>
<td>RFC 9112</td>
</tr>
</tbody>
</table>
</section>
</section>
</middle>
<back>
<displayreference xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
xmlns:x="http://purl.org/net/xml2rfc/ext" target="HTTP10" to="HTTP/1.0"/>
<references>
<name>References</name>
<references>
<name>Normative References</name>
<!-- [HTTP][I-D.ietf-httpbis-semantics]; companion document RFC 9110 -->
<reference anchor="HTTP"><!--included from draft-ietf-httpbis-semantics-latest.xml--> anchor='HTTP' target='https://www.rfc-editor.org/info/rfc9110'>
<front>
<title>HTTP Semantics</title>
<author fullname="Roy initials='R' surname='Fielding' 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>
<postalLine>United States of America</postalLine>
</postal>
<email>fielding@gbiv.com</email>
<uri>https://roy.gbiv.com/</uri>
</address> Fielding' role='editor'>
<organization />
</author>
<author fullname="Mark Nottingham"
initials="M."
surname="Nottingham"
role="editor">
<organization>Fastly</organization>
<address>
<postal>
<postalLine>Prahran VIC</postalLine>
<postalLine>Australia</postalLine>
</postal>
<email>mnot@mnot.net</email>
<uri>https://www.mnot.net/</uri>
</address> initials='M' surname='Nottingham' fullname='Mark Nottingham' role='editor'>
<organization />
</author>
<author fullname="Julian Reschke"
initials="J."
surname="Reschke"
role="editor"> initials='J' surname='Reschke' fullname='Julian Reschke' role='editor'>
<organization abbrev="greenbytes">greenbytes GmbH</organization>
<address>
<postal>
<postalLine>Hafenweg 16</postalLine>
<postalLine>48155 Münster</postalLine>
<postalLine>Germany</postalLine>
</postal>
<email>julian.reschke@greenbytes.de</email>
<uri>https://greenbytes.de/tech/webdav/</uri>
</address> />
</author>
<date year="2021" month="September" day="10"/> year='2022' month='June'/>
</front>
<seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-semantics-19"/> name="STD" value="97"/>
<seriesInfo name="RFC" value="9110"/>
<seriesInfo name="DOI" value="10.17487/RFC9110"/>
</reference>
<!-- [CACHING][I-D.ietf-httpbis-cache]; companion document RFC 9111 -->
<reference anchor="CACHING"><!--included from draft-ietf-httpbis-cache-latest.xml--> anchor='CACHING' target='https://www.rfc-editor.org/info/rfc9111'>
<front>
<title>HTTP Caching</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>
<postalLine>United States of America</postalLine>
</postal>
<email>fielding@gbiv.com</email>
<uri>https://roy.gbiv.com/</uri>
</address> initials='R' surname='Fielding' fullname='Roy Fielding' role='editor'>
<organization />
</author>
<author fullname="Mark Nottingham"
initials="M."
surname="Nottingham"
role="editor">
<organization>Fastly</organization>
<address>
<postal>
<postalLine>Prahran VIC</postalLine>
<postalLine>Australia</postalLine>
</postal>
<email>mnot@mnot.net</email>
<uri>https://www.mnot.net/</uri>
</address> initials='M' surname='Nottingham' fullname='Mark Nottingham' role='editor'>
<organization />
</author>
<author fullname="Julian Reschke"
initials="J."
surname="Reschke"
role="editor"> initials='J' surname='Reschke' fullname='Julian Reschke' role='editor'>
<organization abbrev="greenbytes">greenbytes GmbH</organization>
<address>
<postal>
<postalLine>Hafenweg 16</postalLine>
<postalLine>48155 Münster</postalLine>
<postalLine>Germany</postalLine>
</postal>
<email>julian.reschke@greenbytes.de</email>
<uri>https://greenbytes.de/tech/webdav/</uri>
</address> />
</author>
<date year="2021" month="September" day="10"/> year='2022' month='June'/>
</front>
<seriesInfo name="Internet-Draft" value="draft-ietf-httpbis-cache-19"/> name="STD" value="98"/>
<seriesInfo name="RFC" value="9111"/>
<seriesInfo name="DOI" value="10.17487/RFC9111"/>
</reference>
<reference anchor="URI" target="https://www.rfc-editor.org/info/rfc3986">
<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"/>
</front>
<seriesInfo name="STD" value="66"/>
<seriesInfo name="RFC" value="3986"/>
<seriesInfo name="DOI" value="10.17487/RFC3986"/>
</reference>
<reference anchor="RFC5234" target="https://www.rfc-editor.org/info/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"/>
<seriesInfo name="DOI" value="10.17487/RFC5234"/>
</reference>
<reference anchor="RFC7405" target="https://www.rfc-editor.org/info/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"/>
<seriesInfo name="DOI" value="10.17487/RFC7405"/>
</reference>
<reference anchor="RFC2119" target="https://www.rfc-editor.org/info/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"/>
<seriesInfo name="DOI" value="10.17487/RFC2119"/>
</reference>
<reference anchor="RFC8174" target="https://www.rfc-editor.org/info/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"/>
<seriesInfo name="DOI" value="10.17487/RFC8174"/>
</reference>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5234.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7405.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2119.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8174.xml"/>
<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" target="https://www.rfc-editor.org/info/rfc1950">
<front>
<title>ZLIB Compressed Data Format Specification version 3.3</title>
<author initials="L.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"/>
<seriesInfo name="DOI" value="10.17487/RFC1950"/>
</reference>
<reference anchor="RFC1951" target="https://www.rfc-editor.org/info/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"/>
<seriesInfo name="DOI" value="10.17487/RFC1951"/>
</reference>
<reference anchor="RFC1952" target="https://www.rfc-editor.org/info/rfc1952">
<front>
<title>GZIP file format specification version 4.3</title>
<author initials="P." surname="Deutsch" fullname="L. Peter Deutsch"/>
<author initials="J-L." surname="Gailly" fullname="Jean-Loup Gailly"/>
<author initials="M." surname="Adler" fullname="Mark Adler"/>
<author initials="L.P." surname="Deutsch" fullname="L. Peter Deutsch"/>
<author initials="G."
surname="Randers-Pehrson"
fullname="Glenn Randers-Pehrson"/>
<date month="May" year="1996"/>
</front>
<seriesInfo name="RFC" value="1952"/>
<seriesInfo name="DOI" value="10.17487/RFC1952"/>
</reference>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.1950.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.1951.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.1952.xml"/>
<reference anchor="TLS13" target="https://www.rfc-editor.org/info/rfc8446">
<front>
<title>The Transport Layer Security (TLS) Protocol Version 1.3</title>
<author initials="E." surname="Rescorla" fullname="E. fullname="Eric Rescorla"/>
<date year="2018" month="August"/>
</front>
<seriesInfo name="RFC" value="8446"/>
<seriesInfo name="DOI" value="10.17487/RFC8446"/>
</reference>
<reference anchor="Welch"> anchor="Welch" target="https://ieeexplore.ieee.org/document/1659158/">
<front>
<title>A Technique for High-Performance Data Compression</title>
<author initials="T. A." initials="T." surname="Welch" fullname="Terry A. Welch"/>
<date month="June" year="1984"/>
</front>
<seriesInfo name="IEEE Computer" value="17(6)"/> name="DOI" value="10.1109/MC.1984.1659158"/>
<refcontent>IEEE Computer 17(6)</refcontent>
</reference>
</references>
<references>
<name>Informative References</name>
<reference anchor="HTTP10" target="https://www.rfc-editor.org/info/rfc1945">
<front>
<title abbrev="HTTP/1.0">Hypertext Transfer Protocol -- HTTP/1.0</title>
<author initials="T." surname="Berners-Lee" fullname="Tim fullname="T. Berners-Lee"/>
<author initials="R.T." initials="R." surname="Fielding" fullname="Roy T. fullname="R. Fielding"/>
<author initials="H.F." surname="Nielsen" fullname="Henrik Frystyk Nielsen"/> initials="H." surname="Frystyk" fullname="H. Frystyk"/>
<date month="May" year="1996"/>
</front>
<seriesInfo name="RFC" value="1945"/>
<seriesInfo name="DOI" value="10.17487/RFC1945"/>
</reference>
<reference anchor="RFC2045" target="https://www.rfc-editor.org/info/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.S."
surname="Borenstein"
fullname="Nathaniel S. Borenstein"/>
<date month="November" year="1996"/>
</front>
<seriesInfo name="RFC" value="2045"/>
<seriesInfo name="DOI" value="10.17487/RFC2045"/>
</reference>
<reference anchor="RFC2046" target="https://www.rfc-editor.org/info/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"/>
<seriesInfo name="DOI" value="10.17487/RFC2046"/>
</reference>
<reference anchor="RFC2049" target="https://www.rfc-editor.org/info/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.S."
surname="Borenstein"
fullname="Nathaniel S. Borenstein"/>
<date month="November" year="1996"/>
</front>
<seriesInfo name="RFC" value="2049"/>
<seriesInfo name="DOI" value="10.17487/RFC2049"/>
</reference>
<reference anchor="RFC2068" target="https://www.rfc-editor.org/info/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="Nielsen" 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"/>
<seriesInfo name="DOI" value="10.17487/RFC2068"/>
</reference>
<reference anchor="RFC2557" target="https://www.rfc-editor.org/info/rfc2557">
<front>
<title abbrev="MIME Encapsulation of Aggregate Documents">MIME Encapsulation of Aggregate Documents, such as HTML (MHTML)</title>
<author initials="F." surname="Palme" fullname="Jacob Palme"/>
<author initials="A." surname="Hopmann" fullname="Alex Hopmann"/>
<author initials="N." surname="Shelness" fullname="Nick Shelness"/>
<author initials="E." surname="Stefferud" fullname="Einar Stefferud"/>
<date year="1999" month="March"/>
</front>
<seriesInfo name="RFC" value="2557"/>
<seriesInfo name="DOI" value="10.17487/RFC2557"/>
</reference>
<reference anchor="RFC5322" target="https://www.rfc-editor.org/info/rfc5322">
<front>
<title>Internet Message Format</title>
<author initials="P." surname="Resnick" fullname="P. Resnick"/>
<date year="2008" month="October"/>
</front>
<seriesInfo name="RFC" value="5322"/>
<seriesInfo name="DOI" value="10.17487/RFC5322"/>
</reference>
<reference anchor="RFC7230" target="https://www.rfc-editor.org/info/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. F."
surname="Reschke"
fullname="Julian F. Reschke"
role="editor"/>
<date month="June" year="2014"/>
</front>
<seriesInfo name="RFC" value="7230"/>
<seriesInfo name="DOI" value="10.17487/RFC7230"/>
</reference>
<reference anchor="RFC8126" target="https://www.rfc-editor.org/info/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"/>
<seriesInfo name="DOI" value="10.17487/RFC8126"/>
</reference>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2045.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2046.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2049.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2068.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2557.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5322.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7230.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8126.xml"/>
<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>
<reference anchor="Err4667"
quote-title="false"
target="https://www.rfc-editor.org/errata/eid4667">
<front>
<title>Erratum ID 4667</title>
<author>
<organization>RFC Errata</organization>
</author>
<date/>
</front>
<seriesInfo name="RFC" value="7230"/>
</reference>
</references>
</references>
<section anchor="collected.abnf" title="Collected ABNF">
<t>In the collected ABNF below, list rules are expanded as per <xref target="HTTP" section="5.6.1"/>.</t>
<sourcecode type="abnf" name="draft-ietf-httpbis-messaging-latest.parsed-abnf"><![CDATA[BWS name="rfc9112.parsed-abnf"><![CDATA[
BWS = <BWS, see [HTTP], Section 5.6.3>
HTTP-message = start-line CRLF *( field-line CRLF ) CRLF [
message-body ]
HTTP-name = %x48.54.54.50 ; HTTP
HTTP-version = HTTP-name "/" DIGIT "." DIGIT
OWS = <OWS, see [HTTP], Section 5.6.3>
RWS = <RWS, see [HTTP], Section 5.6.3>
Transfer-Encoding = [ transfer-coding *( OWS "," OWS transfer-coding
) ]
absolute-URI = <absolute-URI, see [URI], Section 4.3>
absolute-form = absolute-URI
absolute-path = <absolute-path, see [HTTP], Section 4.1>
asterisk-form = "*"
authority = <authority, see [URI], Section 3.2>
authority-form = uri-host ":" port
chunk = chunk-size [ chunk-ext ] CRLF chunk-data CRLF
chunk-data = 1*OCTET
chunk-ext = *( BWS ";" BWS chunk-ext-name [ BWS "=" BWS chunk-ext-val
] )
chunk-ext-name = token
chunk-ext-val = token / quoted-string
chunk-size = 1*HEXDIG
chunked-body = *chunk last-chunk trailer-section CRLF
field-line = field-name ":" OWS field-value OWS
field-name = <field-name, see [HTTP], Section 5.1>
field-value = <field-value, see [HTTP], Section 5.5>
last-chunk = 1*"0" [ chunk-ext ] CRLF
message-body = *OCTET
method = token
obs-fold = OWS CRLF RWS
obs-text = <obs-text, see [HTTP], Section 5.6.4>
origin-form = absolute-path [ "?" query ]
port = <port, see [URI], Section 3.2.3>
query = <query, see [URI], Section 3.4>
quoted-string = <quoted-string, see [HTTP], Section 5.6.4>
reason-phrase = 1*( HTAB / SP / VCHAR / obs-text )
request-line = method SP request-target SP HTTP-version
request-target = origin-form / absolute-form / authority-form /
asterisk-form
start-line = request-line / status-line
status-code = 3DIGIT
status-line = HTTP-version SP status-code SP [ reason-phrase ]
token = <token, see [HTTP], Section 5.6.2>
trailer-section = *( field-line CRLF )
transfer-coding = <transfer-coding, see [HTTP], Section 10.1.4>
uri-host = <host, see [URI], Section 3.2.2>
]]></sourcecode>
</section>
<section anchor="differences.between.http.and.mime"
title="Differences between HTTP and MIME">
<t>
HTTP/1.1 uses many of the constructs defined for the
Internet Message Format <xref target="RFC5322"/> and the 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, RFC 2045
is focused only on email; applications some
of HTTP these constructs have many characteristics
that differ from email; hence, HTTP has features that differ from MIME. 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, to allow
greater freedom in the use of new media types, to
make ease date comparisons easier, comparisons,
and to acknowledge the practice of some early
HTTP servers and clients. 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 anchor="mime-version" title="MIME-Version">
<iref primary="true" item="Fields" subitem="MIME-Version"/>
<iref primary="true" item="Header Fields" subitem="MIME-Version"/>
<iref primary="true" item="MIME-Version header field"/>
<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 anchor="conversion.to.canonical.form"
title="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" section="4"/>, and that content with a type of "text" represent 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 Content-Encoding
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 anchor="conversion.of.date.formats" title="Conversion of Date Formats">
<t>
HTTP/1.1 uses a restricted set of date formats (<xref target="HTTP" section="5.6.7"/>) to
simplify the process of date comparison. Proxies and gateways from
other protocols ought to ensure that any Date header field
present in a message conforms to one of the HTTP/1.1 formats and rewrite
the date if necessary.
</t>
</section>
<section anchor="conversion.of.content-encoding"
title="Conversion of Content-Encoding">
<t>
MIME does not include any concept equivalent to HTTP/1.1's HTTP's
Content-Encoding 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 Content-Type
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=<content-coding>" to perform
a function equivalent to Content-Encoding. However, this parameter is
not part of the MIME standards). standards.)
</t>
</section>
<section anchor="conversion.of.content-transfer-encoding"
title="Conversion of Content-Transfer-Encoding">
<iref item="Content-Transfer-Encoding header field"/>
<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 anchor="mhtml.line.length" title="MHTML and Line Length Limitations">
<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="HTTP" section="14.6"/>),
does not interpret
the content or any MIME header lines that might be contained therein.
</t>
</section>
</section>
<section anchor="changes" title="Changes from previous Previous RFCs">
<section anchor="changes.from.0.9" title="Changes from HTTP/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 Host 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 anchor="changes.from.1.0" title="Changes from HTTP/1.0">
<section anchor="changes.to.simplify.multihomed.web.servers.and.conserve.ip.addresses"
title="Multihomed Web Servers">
<t>
The requirements that clients and servers support the Host
header field (<xref target="HTTP" section="7.2"/>), 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 other established mechanism for
distinguishing the intended server of a request other than the IP address
to which that request was directed. The Host 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 anchor="compatibility.with.http.1.0.persistent.connections"
title="Keep-Alive 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 target="RFC2068" section="19.7.1"/>.
</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
Connection, 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 "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 anchor="introduction.of.transfer-encoding"
title="Introduction of Transfer-Encoding">
<t>
HTTP/1.1 introduces the <xref target="field.transfer-encoding" format="none">Transfer-Encoding</xref> 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 anchor="changes.from.rfc.7230" title="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 <xref target="authority-form" format="none">authority-form</xref> 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>
Required recipients
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, re-introduced (bad) whitespace around
";" and "=". "=" has been reintroduced. Whitespace was removed
in <xref target="RFC7230"/>, but that change was found to break existing
implementations (see <xref target="Err4667"/>).
implementations. (<xref target="chunked.extension"/>)
</t>
<t>
Trailer field semantics now transcend the specifics of chunked encoding. 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>
Disallowed transfer
Transfer coding parameters called "q" are disallowed in order to avoid
conflicts with the use of ranks in the TE header field.
(<xref target="transfer.coding.registry"/>)
</t>
</section>
</section>
<section anchor="change.log" title="Change Log">
<t>This section is to be removed before publishing as an RFC.</t>
<section anchor="changes.since.publication.as.rfc"
title="Between RFC7230 and draft 00">
<t>
The changes were purely editorial:
</t>
<ul>
<li>Change boilerplate and abstract to indicate the "draft" status, and update references to ancestor specifications.</li>
<li>Adjust historical notes.</li>
<li>Update links to sibling specifications.</li>
<li>Replace sections listing changes from RFC 2616 by new empty sections referring to RFC 723x.</li>
<li>Remove acknowledgements specific to RFC 723x.</li>
<li>Move "Acknowledgements" to the very end and make them unnumbered.</li>
</ul>
</section>
<section anchor="changes.since.00" title="Since draft-ietf-httpbis-messaging-00">
<t>
The changes in this draft are editorial, with respect to HTTP as a whole,
to move all core HTTP semantics into <xref target="HTTP"/>:
</t>
<ul>
<li>Moved introduction, architecture, conformance, and ABNF extensions from
<xref target="RFC7230" format="none">RFC 7230 (Messaging)</xref> to
semantics <xref target="HTTP"/>.</li>
<li>Moved discussion of MIME differences from RFC 7231 (Semantics) to
<xref target="differences.between.http.and.mime"/>
since they mostly cover transforming 1.1 messages.</li>
<li>Moved all extensibility tips, registration procedures, and registry
tables from the IANA considerations to normative sections, reducing the
IANA considerations to just instructions that will be removed prior to
publication as an RFC.</li>
</ul>
</section>
<section anchor="changes.since.01" title="Since draft-ietf-httpbis-messaging-01">
<ul>
<li>Cite RFC 8126 instead of RFC 5226 (<eref target="https://github.com/httpwg/http-core/issues/75" brackets="angle"/>)</li>
<li>Resolved erratum 4779, no change needed here (<eref target="https://github.com/httpwg/http-core/issues/87" brackets="angle"/>, <eref target="https://www.rfc-editor.org/errata/eid4779" brackets="angle"/>)</li>
<li>In <xref target="transfer.codings"/>, fixed prose claiming transfer parameters allow bare names (<eref target="https://github.com/httpwg/http-core/issues/88" brackets="angle"/>, <eref target="https://www.rfc-editor.org/errata/eid4839" brackets="angle"/>)</li>
<li>Resolved erratum 4225, no change needed here (<eref target="https://github.com/httpwg/http-core/issues/90" brackets="angle"/>, <eref target="https://www.rfc-editor.org/errata/eid4225" brackets="angle"/>)</li>
<li>Replace "response code" with "response status code" (<eref target="https://github.com/httpwg/http-core/issues/94" brackets="angle"/>, <eref target="https://www.rfc-editor.org/errata/eid4050" brackets="angle"/>)</li>
<li>In <xref target="persistent.connections"/>, clarify statement about HTTP/1.0 keep-alive (<eref target="https://github.com/httpwg/http-core/issues/96" brackets="angle"/>, <eref target="https://www.rfc-editor.org/errata/eid4205" brackets="angle"/>)</li>
<li>In <xref target="chunked.extension"/>, re-introduce (bad) whitespace around ";" and "=" (<eref target="https://github.com/httpwg/http-core/issues/101"
brackets="angle"/>,
<eref target="https://www.rfc-editor.org/errata/eid4667" brackets="angle"/>, <eref target="https://www.rfc-editor.org/errata/eid4825" brackets="angle"/>)</li>
<li>In <xref target="transfer.coding.registry"/>, state that transfer codings should not use parameters named "q" (<eref target="https://github.com/httpwg/http-core/issues/15" brackets="angle"/>, <eref target="https://www.rfc-editor.org/errata/eid4683" brackets="angle"/>)</li>
<li>In <xref target="transfer.codings"/>, mark coding name "trailers" as reserved in the IANA registry (<eref target="https://github.com/httpwg/http-core/issues/108"
brackets="angle"/>)</li>
</ul>
</section>
<section anchor="changes.since.02" title="Since draft-ietf-httpbis-messaging-02">
<ul>
<li>In <xref target="status.line"/>, explain why the reason phrase should be ignored by clients (<eref target="https://github.com/httpwg/http-core/issues/60" brackets="angle"/>).</li>
<li>Add <xref target="associating.response.to.request"/> to explain how request/response correlation is performed (<eref target="https://github.com/httpwg/http-core/issues/145"
brackets="angle"/>)</li>
</ul>
</section>
<section anchor="changes.since.03" title="Since draft-ietf-httpbis-messaging-03">
<ul>
<li>In <xref target="associating.response.to.request"/>, caution against treating data on a connection as part of a not-yet-issued request (<eref target="https://github.com/httpwg/http-core/issues/26" brackets="angle"/>)</li>
<li>In <xref target="transfer.codings"/>, remove the predefined codings from the ABNF and make it generic instead (<eref target="https://github.com/httpwg/http-core/issues/66" brackets="angle"/>)</li>
<li>Use RFC 7405 ABNF notation for case-sensitive string constants (<eref target="https://github.com/httpwg/http-core/issues/133"
brackets="angle"/>)</li>
</ul>
</section>
<section anchor="changes.since.04" title="Since draft-ietf-httpbis-messaging-04">
<ul>
<li>In <xref target="HTTP" section="7.8"/>, clarify that protocol-name is to be matched case-insensitively (<eref target="https://github.com/httpwg/http-core/issues/8" brackets="angle"/>)</li>
<li>In <xref target="line.folding"/>, add leading optional whitespace to obs-fold ABNF (<eref target="https://github.com/httpwg/http-core/issues/19" brackets="angle"/>, <eref target="https://www.rfc-editor.org/errata/eid4189" brackets="angle"/>)</li>
<li>In <xref target="status.line"/>, add clarifications about empty reason phrases (<eref target="https://github.com/httpwg/http-core/issues/197"
brackets="angle"/>)</li>
<li>Move discussion of retries from <xref target="persistent.retrying.requests"/> into <xref target="HTTP"/> (<eref target="https://github.com/httpwg/http-core/issues/230"
brackets="angle"/>)</li>
</ul>
</section>
<section anchor="changes.since.05" title="Since draft-ietf-httpbis-messaging-05">
<ul>
<li>In <xref target="chunked.trailer.section"/>, the trailer part has been renamed the trailer section (for consistency with the header section) and trailers are no longer merged as header fields by default, but rather can be discarded, kept separate from header fields, or merged with header fields only if understood and defined as being mergeable (<eref target="https://github.com/httpwg/http-core/issues/16" brackets="angle"/>)</li>
<li>In <xref target="message.format"/> and related Sections, move the trailing CRLF from the line grammars into the message format (<eref target="https://github.com/httpwg/http-core/issues/62" brackets="angle"/>)</li>
<li>Moved <xref target="http.version"/> down (<eref target="https://github.com/httpwg/http-core/issues/68" brackets="angle"/>)</li>
<li>In <xref target="HTTP" section="7.8"/>, use 'websocket' instead of 'HTTP/2.0' in examples (<eref target="https://github.com/httpwg/http-core/issues/112"
brackets="angle"/>)</li>
<li>Move version non-specific text from <xref target="message.body"/> into semantics as "payload" (<eref target="https://github.com/httpwg/http-core/issues/159"
brackets="angle"/>)</li>
<li>In <xref target="tls.connection.closure"/>, add text from RFC 2818 (<eref target="https://github.com/httpwg/http-core/issues/236"
brackets="angle"/>)</li>
</ul>
</section>
<section anchor="changes.since.06" title="Since draft-ietf-httpbis-messaging-06">
<ul>
<li>In <xref target="alpn.registration"/>, update the ALPN protocol ID for HTTP/1.1 (<eref target="https://github.com/httpwg/http-core/issues/49" brackets="angle"/>)</li>
<li>In <xref target="header.field.syntax"/>, align with updates to field terminology in semantics (<eref target="https://github.com/httpwg/http-core/issues/111"
brackets="angle"/>)</li>
<li>In <xref target="HTTP" section="7.6.1"/>, clarify that new connection options indeed need to be registered (<eref target="https://github.com/httpwg/http-core/issues/285"
brackets="angle"/>)</li>
<li>In <xref target="requirements.notation"/>, reference RFC 8174 as well (<eref target="https://github.com/httpwg/http-core/issues/303"
brackets="angle"/>)</li>
</ul>
</section>
<section anchor="changes.since.07" title="Since draft-ietf-httpbis-messaging-07">
<ul>
<li>Move TE: trailers into <xref target="HTTP"/> (<eref target="https://github.com/httpwg/http-core/issues/18" brackets="angle"/>)</li>
<li>In <xref target="message.body.length"/>, adjust requirements for handling multiple content-length values (<eref target="https://github.com/httpwg/http-core/issues/59" brackets="angle"/>)</li>
<li>Throughout, replace "effective request URI" with "target URI" (<eref target="https://github.com/httpwg/http-core/issues/259"
brackets="angle"/>)</li>
<li>In <xref target="field.transfer-encoding"/>, don't claim Transfer-Encoding is supported by HTTP/2 or later (<eref target="https://github.com/httpwg/http-core/issues/297"
brackets="angle"/>)</li>
</ul>
</section>
<section anchor="changes.since.08" title="Since draft-ietf-httpbis-messaging-08">
<ul>
<li>In <xref target="message.parsing"/>, disallow bare CRs (<eref target="https://github.com/httpwg/http-core/issues/31" brackets="angle"/>)</li>
<li>
<xref target="collected.abnf"/> now uses the sender variant of the "#" list expansion (<eref target="https://github.com/httpwg/http-core/issues/192"
brackets="angle"/>)</li>
<li>In <xref target="header.field.syntax"/>, adjust IANA "Close" entry for new registry format (<eref target="https://github.com/httpwg/http-core/issues/273"
brackets="angle"/>)</li>
</ul>
</section>
<section anchor="changes.since.09" title="Since draft-ietf-httpbis-messaging-09">
<ul>
<li>Switch to xml2rfc v3 mode for draft generation (<eref target="https://github.com/httpwg/http-core/issues/394"
brackets="angle"/>)</li>
</ul>
</section>
<section anchor="changes.since.10" title="Since draft-ietf-httpbis-messaging-10">
<ul>
<li>In <xref target="message.body.length"/>, note that TCP half-close does not delimit a request; talk about corresponding server-side behaviour in <xref target="persistent.tear-down"/> (<eref target="https://github.com/httpwg/http-core/issues/22" brackets="angle"/>)</li>
<li>Moved requirements specific to HTTP/1.1 from <xref target="HTTP"/> into <xref target="request.target"/> (<eref target="https://github.com/httpwg/http-core/issues/182"
brackets="angle"/>)</li>
<li>In <xref target="field.transfer-encoding"/> (<xref target="field.transfer-encoding" format="none">Transfer-Encoding</xref>), adjust ABNF to allow empty lists (<eref target="https://github.com/httpwg/http-core/issues/210"
brackets="angle"/>)</li>
<li>In <xref target="tls.connection.initiation"/>, add text from RFC 2818 (<eref target="https://github.com/httpwg/http-core/issues/236"
brackets="angle"/>)</li>
<li>Moved definitions of "TE" and "Upgrade" into <xref target="HTTP"/> (<eref target="https://github.com/httpwg/http-core/issues/392"
brackets="angle"/>)</li>
<li>Moved definition of "Connection" into <xref target="HTTP"/> (<eref target="https://github.com/httpwg/http-core/issues/407"
brackets="angle"/>)</li>
</ul>
</section>
<section anchor="changes.since.11" title="Since draft-ietf-httpbis-messaging-11">
<ul>
<li>Move IANA Upgrade Token Registry instructions to <xref target="HTTP"/> (<eref target="https://github.com/httpwg/http-core/issues/450"
brackets="angle"/>)</li>
</ul>
</section>
<section anchor="changes.since.12" title="Since draft-ietf-httpbis-messaging-12">
<ul>
<li>Moved content of history appendix to Semantics (<eref target="https://github.com/httpwg/http-core/issues/451"
brackets="angle"/>)</li>
<li>Moved note about "close" being reserved as field name to <xref target="persistent.connections"/> (<eref target="https://github.com/httpwg/http-core/issues/500"
brackets="angle"/>)</li>
<li>Moved table of transfer codings into <xref target="transfer.coding.registration"/> (<eref target="https://github.com/httpwg/http-core/issues/506"
brackets="angle"/>)</li>
<li>In Section 13.2, updated the URI for the <xref target="Linhart"/> paper (<eref target="https://github.com/httpwg/http-core/issues/517"
brackets="angle"/>)</li>
<li>Changed document title to just "HTTP/1.1" (<eref target="https://github.com/httpwg/http-core/issues/524"
brackets="angle"/>)</li>
<li>In <xref target="transfer.codings"/>, moved transfer-coding ABNF to <xref target="HTTP" section="10.1.4"/> (<eref target="https://github.com/httpwg/http-core/issues/531"
brackets="angle"/>)</li>
<li>Changed to using "payload data" when defining requirements about the data being conveyed within a message, instead of the terms "payload body" or "response body" or "representation body", since they often get confused with the HTTP/1.1 message body (which includes transfer coding) (<eref target="https://github.com/httpwg/http-core/issues/553"
brackets="angle"/>)</li>
</ul>
</section>
<section anchor="changes.since.13" title="Since draft-ietf-httpbis-messaging-13">
<ul>
<li>In <xref target="message.body.length"/>, clarify that a message needs to be checked for both Content-Length and Transfer-Encoding, before processing Transfer-Encoding, and that ought to be treated as an error, but an intermediary can choose to forward the message downstream after removing the Content-Length and processing the Transfer-Encoding (<eref target="https://github.com/httpwg/http-core/issues/617"
brackets="angle"/>)</li>
<li>Changed to using "content" instead of "payload" or "payload data" to avoid confusion with the payload of version-specific messaging frames (<eref target="https://github.com/httpwg/http-core/issues/654"
brackets="angle"/>)</li>
</ul>
</section>
<section anchor="changes.since.14" title="Since draft-ietf-httpbis-messaging-14">
<ul>
<li>In <xref target="persistent.tear-down"/>, define the close connection option, since its definition was removed from the Connection header field for being specific to 1.1 (<eref target="https://github.com/httpwg/http-core/issues/678"
brackets="angle"/>)</li>
<li>In <xref target="reconstructing.target.uri"/>, clarify how the target URI is reconstructed when the request-target is not in absolute-form and highlight risk in selecting a default host (<eref target="https://github.com/httpwg/http-core/issues/722"
brackets="angle"/>)</li>
<li>In <xref target="chunked.encoding"/>, clarify large chunk handling issues (<eref target="https://github.com/httpwg/http-core/issues/749"
brackets="angle"/>)</li>
<li>In <xref target="message.parsing"/>, explicitly close the connection after sending a 400 (<eref target="https://github.com/httpwg/http-core/issues/750"
brackets="angle"/>)</li>
<li>In <xref target="http.version"/>, refine version requirements for intermediaries (<eref target="https://github.com/httpwg/http-core/issues/751"
brackets="angle"/>)</li>
<li>In <xref target="decoding.chunked"/>, don't remove the Trailer header field (<eref target="https://github.com/httpwg/http-core/issues/793"
brackets="angle"/>)</li>
<li>In <xref target="authority-form"/>, changed the ABNF definition of authority-form from the authority component (in which port is optional) to the host:port format that has always been required by CONNECT (<eref target="https://github.com/httpwg/http-core/issues/806"
brackets="angle"/>)</li>
</ul>
</section>
<section anchor="changes.since.15" title="Since draft-ietf-httpbis-messaging-15">
<ul>
<li>None.</li>
</ul>
</section>
<section anchor="changes.since.16" title="Since draft-ietf-httpbis-messaging-16">
<t>
This draft addresses mostly editorial issues raised during or past IETF
Last Call; see <eref target="https://github.com/httpwg/http-core/issues?q=label%3Ah1-messaging+created%3A%3E2021-05-26"
brackets="angle"/>
for a summary.
</t>
<t>
Furthermore:
</t>
<ul>
<li>In <xref target="field.transfer-encoding"/>, require recipients to avoid smuggling/splitting attacks when processing an ambiguous message framing (<eref target="https://github.com/httpwg/http-core/issues/879"
brackets="angle"/>)</li>
</ul>
</section>
<section anchor="changes.since.17" title="Since draft-ietf-httpbis-messaging-17">
<ul>
<li>In <xref target="status.line"/>, rephrase text about status code definitions in <xref target="HTTP"/> (<eref target="https://github.com/httpwg/http-core/issues/915"
brackets="angle"/>)</li>
<li>In <xref target="associating.response.to.request"/>, clarify how to match responses to requests (<eref target="https://github.com/httpwg/http-core/issues/915"
brackets="angle"/>)</li>
<li>Made reference to <xref target="RFC5322"/> normative, as it is referenced from the ABNF (for "From" header field) (<eref target="https://github.com/httpwg/http-core/issues/915"
brackets="angle"/>)</li>
<li>In <xref target="line.folding"/>, include text about message/http that previously was in <xref target="HTTP"/> (<eref target="https://github.com/httpwg/http-core/issues/923"
brackets="angle"/>)</li>
<li>Throughout, disambiguate "selected representation" and "selected response" (now "chosen response") (<eref target="https://github.com/httpwg/http-core/issues/958"
brackets="angle"/>)</li>
</ul>
</section>
<section anchor="changes.since.18" title="Since draft-ietf-httpbis-messaging-18">
<ul>
<li>Improve a few crossrefs into <xref target="HTTP"/> (<eref target="https://github.com/httpwg/http-core/issues/966"
brackets="angle"/>)</li>
<li>In <xref target="chunked.trailer.section"/>, improve readability of formerly overlong sentence (<eref target="https://github.com/httpwg/http-core/issues/966"
brackets="angle"/>)</li>
<li>Slightly rephrase <xref target="tls.connection.closure"/> (<eref target="https://github.com/httpwg/http-core/pull/972" brackets="angle"/>)</li>
</ul>
</section>
</section>
<section anchor="acks" numbered="false" title="Acknowledgements">
<t>
See Appendix "Acknowledgements" of <xref target="HTTP"/>. target="HTTP"/>, which applies to this document as well.
</t>
</section>
</back>
</rfc>