blob: 1962ca1ee13ebd67564145c0f6923173f82a64e1 [file] [log] [blame]
<?xml version='1.0' encoding='ISO-8859-1' standalone='no'?>
<!DOCTYPE spec SYSTEM "../valid/dtds/spec.dtd" [
<!-- LAST TOUCHED BY: Tim Bray, 8 February 1997 -->
<!-- The words 'FINAL EDIT' in comments mark places where changes
need to be made after approval of the document by the ERB, before
publication. -->
<!ENTITY XML.version "1.0">
<!ENTITY doc.date "10 February 1998">
<!ENTITY iso6.doc.date "19980210">
<!ENTITY w3c.doc.date "02-Feb-1998">
<!ENTITY draft.day '10'>
<!ENTITY draft.month 'February'>
<!ENTITY draft.year '1998'>
<!ENTITY WebSGML
'WebSGML Adaptations Annex to ISO 8879'>
<!ENTITY lt "<">
<!ENTITY gt ">">
<!ENTITY xmlpio "'&lt;?xml'">
<!ENTITY pic "'?>'">
<!ENTITY br "\n">
<!ENTITY cellback '#c0d9c0'>
<!ENTITY mdash "--"> <!-- &#x2014, but nsgmls doesn't grok hex -->
<!ENTITY com "--">
<!ENTITY como "--">
<!ENTITY comc "--">
<!ENTITY hcro "&amp;#x">
<!-- <!ENTITY nbsp " "> -->
<!ENTITY nbsp "&#160;">
<!ENTITY magicents "<code>amp</code>,
<code>lt</code>,
<code>gt</code>,
<code>apos</code>,
<code>quot</code>">
<!-- audience and distribution status: for use at publication time -->
<!ENTITY doc.audience "public review and discussion">
<!ENTITY doc.distribution "may be distributed freely, as long as
all text and legal notices remain intact">
]>
<!-- for Panorama *-->
<?VERBATIM "eg" ?>
<spec>
<header>
<title>Extensible Markup Language (XML) 1.0</title>
<version></version>
<w3c-designation>REC-xml-&iso6.doc.date;</w3c-designation>
<w3c-doctype>W3C Recommendation</w3c-doctype>
<pubdate><day>&draft.day;</day><month>&draft.month;</month><year>&draft.year;</year></pubdate>
<publoc>
<loc href="http://www.w3.org/TR/1998/REC-xml-&iso6.doc.date;">
http://www.w3.org/TR/1998/REC-xml-&iso6.doc.date;</loc>
<loc href="http://www.w3.org/TR/1998/REC-xml-&iso6.doc.date;.xml">
http://www.w3.org/TR/1998/REC-xml-&iso6.doc.date;.xml</loc>
<loc href="http://www.w3.org/TR/1998/REC-xml-&iso6.doc.date;.html">
http://www.w3.org/TR/1998/REC-xml-&iso6.doc.date;.html</loc>
<loc href="http://www.w3.org/TR/1998/REC-xml-&iso6.doc.date;.pdf">
http://www.w3.org/TR/1998/REC-xml-&iso6.doc.date;.pdf</loc>
<loc href="http://www.w3.org/TR/1998/REC-xml-&iso6.doc.date;.ps">
http://www.w3.org/TR/1998/REC-xml-&iso6.doc.date;.ps</loc>
</publoc>
<latestloc>
<loc href="http://www.w3.org/TR/REC-xml">
http://www.w3.org/TR/REC-xml</loc>
</latestloc>
<prevlocs>
<loc href="http://www.w3.org/TR/PR-xml-971208">
http://www.w3.org/TR/PR-xml-971208</loc>
<!--
<loc href='http://www.w3.org/TR/WD-xml-961114'>
http://www.w3.org/TR/WD-xml-961114</loc>
<loc href='http://www.w3.org/TR/WD-xml-lang-970331'>
http://www.w3.org/TR/WD-xml-lang-970331</loc>
<loc href='http://www.w3.org/TR/WD-xml-lang-970630'>
http://www.w3.org/TR/WD-xml-lang-970630</loc>
<loc href='http://www.w3.org/TR/WD-xml-970807'>
http://www.w3.org/TR/WD-xml-970807</loc>
<loc href='http://www.w3.org/TR/WD-xml-971117'>
http://www.w3.org/TR/WD-xml-971117</loc>-->
</prevlocs>
<authlist>
<author><name>Tim Bray</name>
<affiliation>Textuality and Netscape</affiliation>
<email
href="mailto:tbray@textuality.com">tbray@textuality.com</email></author>
<author><name>Jean Paoli</name>
<affiliation>Microsoft</affiliation>
<email href="mailto:jeanpa@microsoft.com">jeanpa@microsoft.com</email></author>
<author><name>C. M. Sperberg-McQueen</name>
<affiliation>University of Illinois at Chicago</affiliation>
<email href="mailto:cmsmcq@uic.edu">cmsmcq@uic.edu</email></author>
</authlist>
<abstract>
<p>The Extensible Markup Language (XML) is a subset of
SGML that is completely described in this document. Its goal is to
enable generic SGML to be served, received, and processed on the Web
in the way that is now possible with HTML. XML has been designed for
ease of implementation and for interoperability with both SGML and
HTML.</p>
</abstract>
<status>
<p>This document has been reviewed by W3C Members and
other interested parties and has been endorsed by the
Director as a W3C Recommendation. It is a stable
document and may be used as reference material or cited
as a normative reference from another document. W3C's
role in making the Recommendation is to draw attention
to the specification and to promote its widespread
deployment. This enhances the functionality and
interoperability of the Web.</p>
<p>
This document specifies a syntax created by subsetting an existing,
widely used international text processing standard (Standard
Generalized Markup Language, ISO 8879:1986(E) as amended and
corrected) for use on the World Wide Web. It is a product of the W3C
XML Activity, details of which can be found at <loc
href='http://www.w3.org/XML'>http://www.w3.org/XML</loc>. A list of
current W3C Recommendations and other technical documents can be found
at <loc href='http://www.w3.org/TR'>http://www.w3.org/TR</loc>.
</p>
<p>This specification uses the term URI, which is defined by <bibref
ref="Berners-Lee"/>, a work in progress expected to update <bibref
ref="RFC1738"/> and <bibref ref="RFC1808"/>.
</p>
<p>The list of known errors in this specification is
available at
<loc href='http://www.w3.org/XML/xml-19980210-errata'>http://www.w3.org/XML/xml-19980210-errata</loc>.</p>
<p>Please report errors in this document to
<loc href='mailto:xml-editor@w3.org'>xml-editor@w3.org</loc>.
</p>
</status>
<pubstmt>
<p>Chicago, Vancouver, Mountain View, et al.:
World-Wide Web Consortium, XML Working Group, 1996, 1997.</p>
</pubstmt>
<sourcedesc>
<p>Created in electronic form.</p>
</sourcedesc>
<langusage>
<language id='EN'>English</language>
<language id='ebnf'>Extended Backus-Naur Form (formal grammar)</language>
</langusage>
<revisiondesc>
<slist>
<sitem>1997-12-03 : CMSMcQ : yet further changes</sitem>
<sitem>1997-12-02 : TB : further changes (see TB to XML WG,
2 December 1997)</sitem>
<sitem>1997-12-02 : CMSMcQ : deal with as many corrections and
comments from the proofreaders as possible:
entify hard-coded document date in pubdate element,
change expansion of entity WebSGML,
update status description as per Dan Connolly (am not sure
about refernece to Berners-Lee et al.),
add 'The' to abstract as per WG decision,
move Relationship to Existing Standards to back matter and
combine with References,
re-order back matter so normative appendices come first,
re-tag back matter so informative appendices are tagged informdiv1,
remove XXX XXX from list of 'normative' specs in prose,
move some references from Other References to Normative References,
add RFC 1738, 1808, and 2141 to Other References (they are not
normative since we do not require the processor to enforce any
rules based on them),
add reference to 'Fielding draft' (Berners-Lee et al.),
move notation section to end of body,
drop URIchar non-terminal and use SkipLit instead,
lose stray reference to defunct nonterminal 'markupdecls',
move reference to Aho et al. into appendix (Tim's right),
add prose note saying that hash marks and fragment identifiers are
NOT part of the URI formally speaking, and are NOT legal in
system identifiers (processor 'may' signal an error).
Work through:
Tim Bray reacting to James Clark,
Tim Bray on his own,
Eve Maler,
NOT DONE YET:
change binary / text to unparsed / parsed.
handle James's suggestion about &lt; in attriubte values
uppercase hex characters,
namechar list,
</sitem>
<sitem>1997-12-01 : JB : add some column-width parameters</sitem>
<sitem>1997-12-01 : CMSMcQ : begin round of changes to incorporate
recent WG decisions and other corrections:
binding sources of character encoding info (27 Aug / 3 Sept),
correct wording of Faust quotation (restore dropped line),
drop SDD from EncodingDecl,
change text at version number 1.0,
drop misleading (wrong!) sentence about ignorables and extenders,
modify definition of PCData to make bar on msc grammatical,
change grammar's handling of internal subset (drop non-terminal markupdecls),
change definition of includeSect to allow conditional sections,
add integral-declaration constraint on internal subset,
drop misleading / dangerous sentence about relationship of
entities with system storage objects,
change table body tag to htbody as per EM change to DTD,
add rule about space normalization in public identifiers,
add description of how to generate our name-space rules from
Unicode character database (needs further work!).
</sitem>
<sitem>1997-10-08 : TB : Removed %-constructs again, new rules
for PE appearance.</sitem>
<sitem>1997-10-01 : TB : Case-sensitive markup; cleaned up
element-type defs, lotsa little edits for style</sitem>
<sitem>1997-09-25 : TB : Change to elm's new DTD, with
substantial detail cleanup as a side-effect</sitem>
<sitem>1997-07-24 : CMSMcQ : correct error (lost *) in definition
of ignoreSectContents (thanks to Makoto Murata)</sitem>
<sitem>Allow all empty elements to have end-tags, consistent with
SGML TC (as per JJC).</sitem>
<sitem>1997-07-23 : CMSMcQ : pre-emptive strike on pending corrections:
introduce the term 'empty-element tag', note that all empty elements
may use it, and elements declared EMPTY must use it.
Add WFC requiring encoding decl to come first in an entity.
Redefine notations to point to PIs as well as binary entities.
Change autodetection table by removing bytes 3 and 4 from
examples with Byte Order Mark.
Add content model as a term and clarify that it applies to both
mixed and element content.
</sitem>
<sitem>1997-06-30 : CMSMcQ : change date, some cosmetic changes,
changes to productions for choice, seq, Mixed, NotationType,
Enumeration. Follow James Clark's suggestion and prohibit
conditional sections in internal subset. TO DO: simplify
production for ignored sections as a result, since we don't
need to worry about parsers which don't expand PErefs finding
a conditional section.</sitem>
<sitem>1997-06-29 : TB : various edits</sitem>
<sitem>1997-06-29 : CMSMcQ : further changes:
Suppress old FINAL EDIT comments and some dead material.
Revise occurrences of % in grammar to exploit Henry Thompson's pun,
especially markupdecl and attdef.
Remove RMD requirement relating to element content (?).
</sitem>
<sitem>1997-06-28 : CMSMcQ : Various changes for 1 July draft:
Add text for draconian error handling (introduce
the term Fatal Error).
RE deleta est (changing wording from
original announcement to restrict the requirement to validating
parsers).
Tag definition of validating processor and link to it.
Add colon as name character.
Change def of %operator.
Change standard definitions of lt, gt, amp.
Strip leading zeros from #x00nn forms.</sitem>
<sitem>1997-04-02 : CMSMcQ : final corrections of editorial errors
found in last night's proofreading. Reverse course once more on
well-formed: Webster's Second hyphenates it, and that's enough
for me.</sitem>
<sitem>1997-04-01 : CMSMcQ : corrections from JJC, EM, HT, and self</sitem>
<sitem>1997-03-31 : Tim Bray : many changes</sitem>
<sitem>1997-03-29 : CMSMcQ : some Henry Thompson (on entity handling),
some Charles Goldfarb, some ERB decisions (PE handling in miscellaneous
declarations. Changed Ident element to accept def attribute.
Allow normalization of Unicode characters. move def of systemliteral
into section on literals.</sitem>
<sitem>1997-03-28 : CMSMcQ : make as many corrections as possible, from
Terry Allen, Norbert Mikula, James Clark, Jon Bosak, Henry Thompson,
Paul Grosso, and self. Among other things: give in on "well formed"
(Terry is right), tentatively rename QuotedCData as AttValue
and Literal as EntityValue to be more informative, since attribute
values are the <emph>only</emph> place QuotedCData was used, and
vice versa for entity text and Literal. (I'd call it Entity Text,
but 8879 uses that name for both internal and external entities.)</sitem>
<sitem>1997-03-26 : CMSMcQ : resynch the two forks of this draft, reapply
my changes dated 03-20 and 03-21. Normalize old 'may not' to 'must not'
except in the one case where it meant 'may or may not'.</sitem>
<sitem>1997-03-21 : TB : massive changes on plane flight from Chicago
to Vancouver</sitem>
<sitem>1997-03-21 : CMSMcQ : correct as many reported errors as possible.
</sitem>
<sitem>1997-03-20 : CMSMcQ : correct typos listed in CMSMcQ hand copy of spec.</sitem>
<sitem>1997-03-20 : CMSMcQ : cosmetic changes preparatory to revision for
WWW conference April 1997: restore some of the internal entity
references (e.g. to docdate, etc.), change character xA0 to &amp;nbsp;
and define nbsp as &amp;#160;, and refill a lot of paragraphs for
legibility.</sitem>
<sitem>1996-11-12 : CMSMcQ : revise using Tim's edits:
Add list type of NUMBERED and change most lists either to
BULLETS or to NUMBERED.
Suppress QuotedNames, Names (not used).
Correct trivial-grammar doc type decl.
Rename 'marked section' as 'CDATA section' passim.
Also edits from James Clark:
Define the set of characters from which [^abc] subtracts.
Charref should use just [0-9] not Digit.
Location info needs cleaner treatment: remove? (ERB
question).
One example of a PI has wrong pic.
Clarify discussion of encoding names.
Encoding failure should lead to unspecified results; don't
prescribe error recovery.
Don't require exposure of entity boundaries.
Ignore white space in element content.
Reserve entity names of the form u-NNNN.
Clarify relative URLs.
And some of my own:
Correct productions for content model: model cannot
consist of a name, so "elements ::= cp" is no good.
</sitem>
<sitem>1996-11-11 : CMSMcQ : revise for style.
Add new rhs to entity declaration, for parameter entities.</sitem>
<sitem>1996-11-10 : CMSMcQ : revise for style.
Fix / complete section on names, characters.
Add sections on parameter entities, conditional sections.
Still to do: Add compatibility note on deterministic content models.
Finish stylistic revision.</sitem>
<sitem>1996-10-31 : TB : Add Entity Handling section</sitem>
<sitem>1996-10-30 : TB : Clean up term &amp; termdef. Slip in
ERB decision re EMPTY.</sitem>
<sitem>1996-10-28 : TB : Change DTD. Implement some of Michael's
suggestions. Change comments back to //. Introduce language for
XML namespace reservation. Add section on white-space handling.
Lots more cleanup.</sitem>
<sitem>1996-10-24 : CMSMcQ : quick tweaks, implement some ERB
decisions. Characters are not integers. Comments are /* */ not //.
Add bibliographic refs to 10646, HyTime, Unicode.
Rename old Cdata as MsData since it's <emph>only</emph> seen
in marked sections. Call them attribute-value pairs not
name-value pairs, except once. Internal subset is optional, needs
'?'. Implied attributes should be signaled to the app, not
have values supplied by processor.</sitem>
<sitem>1996-10-16 : TB : track down &amp; excise all DSD references;
introduce some EBNF for entity declarations.</sitem>
<sitem>1996-10-?? : TB : consistency check, fix up scraps so
they all parse, get formatter working, correct a few productions.</sitem>
<sitem>1996-10-10/11 : CMSMcQ : various maintenance, stylistic, and
organizational changes:
Replace a few literals with xmlpio and
pic entities, to make them consistent and ensure we can change pic
reliably when the ERB votes.
Drop paragraph on recognizers from notation section.
Add match, exact match to terminology.
Move old 2.2 XML Processors and Apps into intro.
Mention comments, PIs, and marked sections in discussion of
delimiter escaping.
Streamline discussion of doctype decl syntax.
Drop old section of 'PI syntax' for doctype decl, and add
section on partial-DTD summary PIs to end of Logical Structures
section.
Revise DSD syntax section to use Tim's subset-in-a-PI
mechanism.</sitem>
<sitem>1996-10-10 : TB : eliminate name recognizers (and more?)</sitem>
<sitem>1996-10-09 : CMSMcQ : revise for style, consistency through 2.3
(Characters)</sitem>
<sitem>1996-10-09 : CMSMcQ : re-unite everything for convenience,
at least temporarily, and revise quickly</sitem>
<sitem>1996-10-08 : TB : first major homogenization pass</sitem>
<sitem>1996-10-08 : TB : turn "current" attribute on div type into
CDATA</sitem>
<sitem>1996-10-02 : TB : remould into skeleton + entities</sitem>
<sitem>1996-09-30 : CMSMcQ : add a few more sections prior to exchange
with Tim.</sitem>
<sitem>1996-09-20 : CMSMcQ : finish transcribing notes.</sitem>
<sitem>1996-09-19 : CMSMcQ : begin transcribing notes for draft.</sitem>
<sitem>1996-09-13 : CMSMcQ : made outline from notes of 09-06,
do some housekeeping</sitem>
</slist>
</revisiondesc>
</header>
<body>
<div1 id='sec-intro'>
<head>Introduction</head>
<p>Extensible Markup Language, abbreviated XML, describes a class of
data objects called <termref def="dt-xml-doc">XML documents</termref> and
partially describes the behavior of
computer programs which process them. XML is an application profile or
restricted form of SGML, the Standard Generalized Markup
Language <bibref ref='ISO8879'/>.
By construction, XML documents
are conforming SGML documents.
</p>
<p>XML documents are made up of storage units called <termref
def="dt-entity">entities</termref>, which contain either parsed
or unparsed data.
Parsed data is made up of <termref def="dt-character">characters</termref>,
some
of which form <termref def="dt-chardata">character data</termref>,
and some of which form <termref def="dt-markup">markup</termref>.
Markup encodes a description of the document's storage layout and
logical structure. XML provides a mechanism to impose constraints on
the storage layout and logical structure.</p>
<p><termdef id="dt-xml-proc" term="XML Processor">A software module
called an <term>XML processor</term> is used to read XML documents
and provide access to their content and structure.</termdef> <termdef
id="dt-app" term="Application">It is assumed that an XML processor is
doing its work on behalf of another module, called the
<term>application</term>.</termdef> This specification describes the
required behavior of an XML processor in terms of how it must read XML
data and the information it must provide to the application.</p>
<div2 id='sec-origin-goals'>
<head>Origin and Goals</head>
<p>XML was developed by an XML Working Group (originally known as the
SGML Editorial Review Board) formed under the auspices of the World
Wide Web Consortium (W3C) in 1996.
It was chaired by Jon Bosak of Sun
Microsystems with the active participation of an XML Special
Interest Group (previously known as the SGML Working Group) also
organized by the W3C. The membership of the XML Working Group is given
in an appendix. Dan Connolly served as the WG's contact with the W3C.
</p>
<p>The design goals for XML are:<olist>
<item><p>XML shall be straightforwardly usable over the
Internet.</p></item>
<item><p>XML shall support a wide variety of applications.</p></item>
<item><p>XML shall be compatible with SGML.</p></item>
<item><p>It shall be easy to write programs which process XML
documents.</p></item>
<item><p>The number of optional features in XML is to be kept to the
absolute minimum, ideally zero.</p></item>
<item><p>XML documents should be human-legible and reasonably
clear.</p></item>
<item><p>The XML design should be prepared quickly.</p></item>
<item><p>The design of XML shall be formal and concise.</p></item>
<item><p>XML documents shall be easy to create.</p></item>
<item><p>Terseness in XML markup is of minimal importance.</p></item></olist>
</p>
<p>This specification,
together with associated standards
(Unicode and ISO/IEC 10646 for characters,
Internet RFC 1766 for language identification tags,
ISO 639 for language name codes, and
ISO 3166 for country name codes),
provides all the information necessary to understand
XML Version &XML.version;
and construct computer programs to process it.</p>
<p>This version of the XML specification
<!-- is for &doc.audience;.-->
&doc.distribution;.</p>
</div2>
<div2 id='sec-terminology'>
<head>Terminology</head>
<p>The terminology used to describe XML documents is defined in the body of
this specification.
The terms defined in the following list are used in building those
definitions and in describing the actions of an XML processor:
<glist>
<gitem>
<label>may</label>
<def><p><termdef id="dt-may" term="May">Conforming documents and XML
processors are permitted to but need not behave as
described.</termdef></p></def>
</gitem>
<gitem>
<label>must</label>
<def><p>Conforming documents and XML processors
are required to behave as described; otherwise they are in error.
<!-- do NOT change this! this is what defines a violation of
a 'must' clause as 'an error'. -MSM -->
</p></def>
</gitem>
<gitem>
<label>error</label>
<def><p><termdef id='dt-error' term='Error'
>A violation of the rules of this
specification; results are
undefined. Conforming software may detect and report an error and may
recover from it.</termdef></p></def>
</gitem>
<gitem>
<label>fatal error</label>
<def><p><termdef id="dt-fatal" term="Fatal Error">An error
which a conforming <termref def="dt-xml-proc">XML processor</termref>
must detect and report to the application.
After encountering a fatal error, the
processor may continue
processing the data to search for further errors and may report such
errors to the application. In order to support correction of errors,
the processor may make unprocessed data from the document (with
intermingled character data and markup) available to the application.
Once a fatal error is detected, however, the processor must not
continue normal processing (i.e., it must not
continue to pass character data and information about the document's
logical structure to the application in the normal way).
</termdef></p></def>
</gitem>
<gitem>
<label>at user option</label>
<def><p>Conforming software may or must (depending on the modal verb in the
sentence) behave as described; if it does, it must
provide users a means to enable or disable the behavior
described.</p></def>
</gitem>
<gitem>
<label>validity constraint</label>
<def><p>A rule which applies to all
<termref def="dt-valid">valid</termref> XML documents.
Violations of validity constraints are errors; they must, at user option,
be reported by
<termref def="dt-validating">validating XML processors</termref>.</p></def>
</gitem>
<gitem>
<label>well-formedness constraint</label>
<def><p>A rule which applies to all <termref
def="dt-wellformed">well-formed</termref> XML documents.
Violations of well-formedness constraints are
<termref def="dt-fatal">fatal errors</termref>.</p></def>
</gitem>
<gitem>
<label>match</label>
<def><p><termdef id="dt-match" term="match">(Of strings or names:)
Two strings or names being compared must be identical.
Characters with multiple possible representations in ISO/IEC 10646 (e.g.
characters with
both precomposed and base+diacritic forms) match only if they have the
same representation in both strings.
At user option, processors may normalize such characters to
some canonical form.
No case folding is performed.
(Of strings and rules in the grammar:)
A string matches a grammatical production if it belongs to the
language generated by that production.
(Of content and content models:)
An element matches its declaration when it conforms
in the fashion described in the constraint
<specref ref='elementvalid'/>.
</termdef>
</p></def>
</gitem>
<gitem>
<label>for compatibility</label>
<def><p><termdef id="dt-compat" term="For Compatibility">A feature of
XML included solely to ensure that XML remains compatible with SGML.
</termdef></p></def>
</gitem>
<gitem>
<label>for interoperability</label>
<def><p><termdef id="dt-interop" term="For interoperability">A
non-binding recommendation included to increase the chances that XML
documents can be processed by the existing installed base of SGML
processors which predate the
&WebSGML;.</termdef></p></def>
</gitem>
</glist>
</p>
</div2>
</div1>
<!-- &Docs; -->
<div1 id='sec-documents'>
<head>Documents</head>
<p><termdef id="dt-xml-doc" term="XML Document">
A data object is an
<term>XML document</term> if it is
<termref def="dt-wellformed">well-formed</termref>, as
defined in this specification.
A well-formed XML document may in addition be
<termref def="dt-valid">valid</termref> if it meets certain further
constraints.</termdef></p>
<p>Each XML document has both a logical and a physical structure.
Physically, the document is composed of units called <termref
def="dt-entity">entities</termref>. An entity may <termref
def="dt-entref">refer</termref> to other entities to cause their
inclusion in the document. A document begins in a "root" or <termref
def="dt-docent">document entity</termref>.
Logically, the document is composed of declarations, elements,
comments,
character references, and
processing
instructions, all of which are indicated in the document by explicit
markup.
The logical and physical structures must nest properly, as described
in <specref ref='wf-entities'/>.
</p>
<div2 id='sec-well-formed'>
<head>Well-Formed XML Documents</head>
<p><termdef id="dt-wellformed" term="Well-Formed">
A textual object is
a well-formed XML document if:</termdef>
<olist>
<item><p>Taken as a whole, it
matches the production labeled <nt def='NT-document'>document</nt>.</p></item>
<item><p>It
meets all the well-formedness constraints given in this specification.</p>
</item>
<item><p>Each of the <termref def='dt-parsedent'>parsed entities</termref>
which is referenced directly or indirectly within the document is
<titleref href='wf-entities'>well-formed</titleref>.</p></item>
</olist></p>
<p>
<scrap lang='ebnf' id='document'>
<head>Document</head>
<prod id='NT-document'><lhs>document</lhs>
<rhs><nt def='NT-prolog'>prolog</nt>
<nt def='NT-element'>element</nt>
<nt def='NT-Misc'>Misc</nt>*</rhs></prod>
</scrap>
</p>
<p>Matching the <nt def="NT-document">document</nt> production
implies that:
<olist>
<item><p>It contains one or more
<termref def="dt-element">elements</termref>.</p>
</item>
<!--* N.B. some readers (notably JC) find the following
paragraph awkward and redundant. I agree it's logically redundant:
it *says* it is summarizing the logical implications of
matching the grammar, and that means by definition it's
logically redundant. I don't think it's rhetorically
redundant or unnecessary, though, so I'm keeping it. It
could however use some recasting when the editors are feeling
stronger. -MSM *-->
<item><p><termdef id="dt-root" term="Root Element">There is exactly
one element, called the <term>root</term>, or document element, no
part of which appears in the <termref
def="dt-content">content</termref> of any other element.</termdef>
For all other elements, if the start-tag is in the content of another
element, the end-tag is in the content of the same element. More
simply stated, the elements, delimited by start- and end-tags, nest
properly within each other.
</p></item>
</olist>
</p>
<p><termdef id="dt-parentchild" term="Parent/Child">As a consequence
of this,
for each non-root element
<code>C</code> in the document, there is one other element <code>P</code>
in the document such that
<code>C</code> is in the content of <code>P</code>, but is not in
the content of any other element that is in the content of
<code>P</code>.
<code>P</code> is referred to as the
<term>parent</term> of <code>C</code>, and <code>C</code> as a
<term>child</term> of <code>P</code>.</termdef></p></div2>
<div2 id="charsets">
<head>Characters</head>
<p><termdef id="dt-text" term="Text">A parsed entity contains
<term>text</term>, a sequence of
<termref def="dt-character">characters</termref>,
which may represent markup or character data.</termdef>
<termdef id="dt-character" term="Character">A <term>character</term>
is an atomic unit of text as specified by
ISO/IEC 10646 <bibref ref="ISO10646"/>.
Legal characters are tab, carriage return, line feed, and the legal
graphic characters of Unicode and ISO/IEC 10646.
The use of "compatibility characters", as defined in section 6.8
of <bibref ref='Unicode'/>, is discouraged.
</termdef>
<scrap lang="ebnf" id="char32">
<head>Character Range</head>
<prodgroup pcw2="4" pcw4="17.5" pcw5="11">
<prod id="NT-Char"><lhs>Char</lhs>
<rhs>#x9 | #xA | #xD | [#x20-#xD7FF] | [#xE000-#xFFFD]
| [#x10000-#x10FFFF]</rhs>
<com>any Unicode character, excluding the
surrogate blocks, FFFE, and FFFF.</com> </prod>
</prodgroup>
</scrap>
</p>
<p>The mechanism for encoding character code points into bit patterns may
vary from entity to entity. All XML processors must accept the UTF-8
and UTF-16 encodings of 10646; the mechanisms for signaling which of
the two is in use, or for bringing other encodings into play, are
discussed later, in <specref ref='charencoding'/>.
</p>
<!--
<p>Regardless of the specific encoding used, any character in the ISO/IEC
10646 character set may be referred to by the decimal or hexadecimal
equivalent of its
UCS-4 code value.
</p>-->
</div2>
<div2 id='sec-common-syn'>
<head>Common Syntactic Constructs</head>
<p>This section defines some symbols used widely in the grammar.</p>
<p><nt def="NT-S">S</nt> (white space) consists of one or more space (#x20)
characters, carriage returns, line feeds, or tabs.
<scrap lang="ebnf" id='white'>
<head>White Space</head>
<prodgroup pcw2="4" pcw4="17.5" pcw5="11">
<prod id='NT-S'><lhs>S</lhs>
<rhs>(#x20 | #x9 | #xD | #xA)+</rhs>
</prod>
</prodgroup>
</scrap></p>
<p>Characters are classified for convenience as letters, digits, or other
characters. Letters consist of an alphabetic or syllabic
base character possibly
followed by one or more combining characters, or of an ideographic
character.
Full definitions of the specific characters in each class
are given in <specref ref='CharClasses'/>.</p>
<p><termdef id="dt-name" term="Name">A <term>Name</term> is a token
beginning with a letter or one of a few punctuation characters, and continuing
with letters, digits, hyphens, underscores, colons, or full stops, together
known as name characters.</termdef>
Names beginning with the string "<code>xml</code>", or any string
which would match <code>(('X'|'x') ('M'|'m') ('L'|'l'))</code>, are
reserved for standardization in this or future versions of this
specification.
</p>
<note>
<p>The colon character within XML names is reserved for experimentation with
name spaces.
Its meaning is expected to be
standardized at some future point, at which point those documents
using the colon for experimental purposes may need to be updated.
(There is no guarantee that any name-space mechanism
adopted for XML will in fact use the colon as a name-space delimiter.)
In practice, this means that authors should not use the colon in XML
names except as part of name-space experiments, but that XML processors
should accept the colon as a name character.</p>
</note>
<p>An
<nt def='NT-Nmtoken'>Nmtoken</nt> (name token) is any mixture of
name characters.
<scrap lang='ebnf'>
<head>Names and Tokens</head>
<prod id='NT-NameChar'><lhs>NameChar</lhs>
<rhs><nt def="NT-Letter">Letter</nt>
| <nt def='NT-Digit'>Digit</nt>
| '.' | '-' | '_' | ':'
| <nt def='NT-CombiningChar'>CombiningChar</nt>
| <nt def='NT-Extender'>Extender</nt></rhs>
</prod>
<prod id='NT-Name'><lhs>Name</lhs>
<rhs>(<nt def='NT-Letter'>Letter</nt> | '_' | ':')
(<nt def='NT-NameChar'>NameChar</nt>)*</rhs></prod>
<prod id='NT-Names'><lhs>Names</lhs>
<rhs><nt def='NT-Name'>Name</nt>
(<nt def='NT-S'>S</nt> <nt def='NT-Name'>Name</nt>)*</rhs></prod>
<prod id='NT-Nmtoken'><lhs>Nmtoken</lhs>
<rhs>(<nt def='NT-NameChar'>NameChar</nt>)+</rhs></prod>
<prod id='NT-Nmtokens'><lhs>Nmtokens</lhs>
<rhs><nt def='NT-Nmtoken'>Nmtoken</nt> (<nt def='NT-S'>S</nt> <nt def='NT-Nmtoken'>Nmtoken</nt>)*</rhs></prod>
</scrap>
</p>
<p>Literal data is any quoted string not containing
the quotation mark used as a delimiter for that string.
Literals are used
for specifying the content of internal entities
(<nt def='NT-EntityValue'>EntityValue</nt>),
the values of attributes (<nt def='NT-AttValue'>AttValue</nt>),
and external identifiers
(<nt def="NT-SystemLiteral">SystemLiteral</nt>).
Note that a <nt def='NT-SystemLiteral'>SystemLiteral</nt>
can be parsed without scanning for markup.
<scrap lang='ebnf'>
<head>Literals</head>
<prod id='NT-EntityValue'><lhs>EntityValue</lhs>
<rhs>'"'
([^%&amp;"]
| <nt def='NT-PEReference'>PEReference</nt>
| <nt def='NT-Reference'>Reference</nt>)*
'"'
</rhs>
<rhs>|&nbsp;
"'"
([^%&amp;']
| <nt def='NT-PEReference'>PEReference</nt>
| <nt def='NT-Reference'>Reference</nt>)*
"'"</rhs>
</prod>
<prod id='NT-AttValue'><lhs>AttValue</lhs>
<rhs>'"'
([^&lt;&amp;"]
| <nt def='NT-Reference'>Reference</nt>)*
'"'
</rhs>
<rhs>|&nbsp;
"'"
([^&lt;&amp;']
| <nt def='NT-Reference'>Reference</nt>)*
"'"</rhs>
</prod>
<prod id="NT-SystemLiteral"><lhs>SystemLiteral</lhs>
<rhs>('"' [^"]* '"') |&nbsp;("'" [^']* "'")
</rhs>
</prod>
<prod id="NT-PubidLiteral"><lhs>PubidLiteral</lhs>
<rhs>'"' <nt def='NT-PubidChar'>PubidChar</nt>*
'"'
| "'" (<nt def='NT-PubidChar'>PubidChar</nt> - "'")* "'"</rhs>
</prod>
<prod id="NT-PubidChar"><lhs>PubidChar</lhs>
<rhs>#x20 | #xD | #xA
|&nbsp;[a-zA-Z0-9]
|&nbsp;[-'()+,./:=?;!*#@$_%]</rhs>
</prod>
</scrap>
</p>
</div2>
<div2 id='syntax'>
<head>Character Data and Markup</head>
<p><termref def='dt-text'>Text</termref> consists of intermingled
<termref def="dt-chardata">character
data</termref> and markup.
<termdef id="dt-markup" term="Markup"><term>Markup</term> takes the form of
<termref def="dt-stag">start-tags</termref>,
<termref def="dt-etag">end-tags</termref>,
<termref def="dt-empty">empty-element tags</termref>,
<termref def="dt-entref">entity references</termref>,
<termref def="dt-charref">character references</termref>,
<termref def="dt-comment">comments</termref>,
<termref def="dt-cdsection">CDATA section</termref> delimiters,
<termref def="dt-doctype">document type declarations</termref>, and
<termref def="dt-pi">processing instructions</termref>.
</termdef>
</p>
<p><termdef id="dt-chardata" term="Character Data">All text that is not markup
constitutes the <term>character data</term> of
the document.</termdef></p>
<p>The ampersand character (&amp;) and the left angle bracket (&lt;)
may appear in their literal form <emph>only</emph> when used as markup
delimiters, or within a <termref def="dt-comment">comment</termref>, a
<termref def="dt-pi">processing instruction</termref>,
or a <termref def="dt-cdsection">CDATA section</termref>.
They are also legal within the <termref def='dt-litentval'>literal entity
value</termref> of an internal entity declaration; see
<specref ref='wf-entities'/>.
<!-- FINAL EDIT: restore internal entity decl or leave it out. -->
If they are needed elsewhere,
they must be <termref def="dt-escape">escaped</termref>
using either <termref def='dt-charref'>numeric character references</termref>
or the strings
"<code>&amp;amp;</code>" and "<code>&amp;lt;</code>" respectively.
The right angle
bracket (>) may be represented using the string
"<code>&amp;gt;</code>", and must, <termref def='dt-compat'>for
compatibility</termref>,
be escaped using
"<code>&amp;gt;</code>" or a character reference
when it appears in the string
"<code>]]&gt;</code>"
in content,
when that string is not marking the end of
a <termref def="dt-cdsection">CDATA section</termref>.
</p>
<p>
In the content of elements, character data
is any string of characters which does
not contain the start-delimiter of any markup.
In a CDATA section, character data
is any string of characters not including the CDATA-section-close
delimiter, "<code>]]&gt;</code>".</p>
<p>
To allow attribute values to contain both single and double quotes, the
apostrophe or single-quote character (') may be represented as
"<code>&amp;apos;</code>", and the double-quote character (") as
"<code>&amp;quot;</code>".
<scrap lang="ebnf">
<head>Character Data</head>
<prod id='NT-CharData'>
<lhs>CharData</lhs>
<rhs>[^&lt;&amp;]* - ([^&lt;&amp;]* ']]&gt;' [^&lt;&amp;]*)</rhs>
</prod>
</scrap>
</p>
</div2>
<div2 id='sec-comments'>
<head>Comments</head>
<p><termdef id="dt-comment" term="Comment"><term>Comments</term> may
appear anywhere in a document outside other
<termref def='dt-markup'>markup</termref>; in addition,
they may appear within the document type declaration
at places allowed by the grammar.
They are not part of the document's <termref def="dt-chardata">character
data</termref>; an XML
processor may, but need not, make it possible for an application to
retrieve the text of comments.
<termref def="dt-compat">For compatibility</termref>, the string
"<code>--</code>" (double-hyphen) must not occur within
comments.
<scrap lang="ebnf">
<head>Comments</head>
<prod id='NT-Comment'><lhs>Comment</lhs>
<rhs>'&lt;!--'
((<nt def='NT-Char'>Char</nt> - '-')
| ('-' (<nt def='NT-Char'>Char</nt> - '-')))*
'-->'</rhs>
</prod>
</scrap>
</termdef></p>
<p>An example of a comment:
<eg>&lt;!&como; declarations for &lt;head> &amp; &lt;body> &comc;&gt;</eg>
</p>
</div2>
<div2 id='sec-pi'>
<head>Processing Instructions</head>
<p><termdef id="dt-pi" term="Processing instruction"><term>Processing
instructions</term> (PIs) allow documents to contain instructions
for applications.
<scrap lang="ebnf">
<head>Processing Instructions</head>
<prod id='NT-PI'><lhs>PI</lhs>
<rhs>'&lt;?' <nt def='NT-PITarget'>PITarget</nt>
(<nt def='NT-S'>S</nt>
(<nt def='NT-Char'>Char</nt>* -
(<nt def='NT-Char'>Char</nt>* &pic; <nt def='NT-Char'>Char</nt>*)))?
&pic;</rhs></prod>
<prod id='NT-PITarget'><lhs>PITarget</lhs>
<rhs><nt def='NT-Name'>Name</nt> -
(('X' | 'x') ('M' | 'm') ('L' | 'l'))</rhs>
</prod>
</scrap></termdef>
PIs are not part of the document's <termref def="dt-chardata">character
data</termref>, but must be passed through to the application. The
PI begins with a target (<nt def='NT-PITarget'>PITarget</nt>) used
to identify the application to which the instruction is directed.
The target names "<code>XML</code>", "<code>xml</code>", and so on are
reserved for standardization in this or future versions of this
specification.
The
XML <termref def='dt-notation'>Notation</termref> mechanism
may be used for
formal declaration of PI targets.
</p>
</div2>
<div2 id='sec-cdata-sect'>
<head>CDATA Sections</head>
<p><termdef id="dt-cdsection" term="CDATA Section"><term>CDATA sections</term>
may occur
anywhere character data may occur; they are
used to escape blocks of text containing characters which would
otherwise be recognized as markup. CDATA sections begin with the
string "<code>&lt;![CDATA[</code>" and end with the string
"<code>]]&gt;</code>":
<scrap lang="ebnf">
<head>CDATA Sections</head>
<prod id='NT-CDSect'><lhs>CDSect</lhs>
<rhs><nt def='NT-CDStart'>CDStart</nt>
<nt def='NT-CData'>CData</nt>
<nt def='NT-CDEnd'>CDEnd</nt></rhs></prod>
<prod id='NT-CDStart'><lhs>CDStart</lhs>
<rhs>'&lt;![CDATA['</rhs>
</prod>
<prod id='NT-CData'><lhs>CData</lhs>
<rhs>(<nt def='NT-Char'>Char</nt>* -
(<nt def='NT-Char'>Char</nt>* ']]&gt;' <nt def='NT-Char'>Char</nt>*))
</rhs>
</prod>
<prod id='NT-CDEnd'><lhs>CDEnd</lhs>
<rhs>']]&gt;'</rhs>
</prod>
</scrap>
Within a CDATA section, only the <nt def='NT-CDEnd'>CDEnd</nt> string is
recognized as markup, so that left angle brackets and ampersands may occur in
their literal form; they need not (and cannot) be escaped using
"<code>&amp;lt;</code>" and "<code>&amp;amp;</code>". CDATA sections
cannot nest.</termdef>
</p>
<p>An example of a CDATA section, in which "<code>&lt;greeting></code>" and
"<code>&lt;/greeting></code>"
are recognized as <termref def='dt-chardata'>character data</termref>, not
<termref def='dt-markup'>markup</termref>:
<eg>&lt;![CDATA[&lt;greeting>Hello, world!&lt;/greeting>]]&gt;</eg>
</p>
</div2>
<div2 id='sec-prolog-dtd'>
<head>Prolog and Document Type Declaration</head>
<p><termdef id='dt-xmldecl' term='XML Declaration'>XML documents
may, and should,
begin with an <term>XML declaration</term> which specifies
the version of
XML being used.</termdef>
For example, the following is a complete XML document, <termref
def="dt-wellformed">well-formed</termref> but not
<termref def="dt-valid">valid</termref>:
<eg><![CDATA[<?xml version="1.0"?>
<greeting>Hello, world!</greeting>
]]></eg>
and so is this:
<eg><![CDATA[<greeting>Hello, world!</greeting>
]]></eg>
</p>
<p>The version number "<code>1.0</code>" should be used to indicate
conformance to this version of this specification; it is an error
for a document to use the value "<code>1.0</code>"
if it does not conform to this version of this specification.
It is the intent
of the XML working group to give later versions of this specification
numbers other than "<code>1.0</code>", but this intent does not
indicate a
commitment to produce any future versions of XML, nor if any are produced, to
use any particular numbering scheme.
Since future versions are not ruled out, this construct is provided
as a means to allow the possibility of automatic version recognition, should
it become necessary.
Processors may signal an error if they receive documents labeled with
versions they do not support.
</p>
<p>The function of the markup in an XML document is to describe its
storage and logical structure and to associate attribute-value pairs
with its logical structures. XML provides a mechanism, the <termref
def="dt-doctype">document type declaration</termref>, to define
constraints on the logical structure and to support the use of
predefined storage units.
<termdef id="dt-valid" term="Validity">An XML document is
<term>valid</term> if it has an associated document type
declaration and if the document
complies with the constraints expressed in it.</termdef></p>
<p>The document type declaration must appear before
the first <termref def="dt-element">element</termref> in the document.
<scrap lang="ebnf" id='xmldoc'>
<head>Prolog</head>
<prodgroup pcw2="6" pcw4="17.5" pcw5="9">
<prod id='NT-prolog'><lhs>prolog</lhs>
<rhs><nt def='NT-XMLDecl'>XMLDecl</nt>?
<nt def='NT-Misc'>Misc</nt>*
(<nt def='NT-doctypedecl'>doctypedecl</nt>
<nt def='NT-Misc'>Misc</nt>*)?</rhs></prod>
<prod id='NT-XMLDecl'><lhs>XMLDecl</lhs>
<rhs>&xmlpio;
<nt def='NT-VersionInfo'>VersionInfo</nt>
<nt def='NT-EncodingDecl'>EncodingDecl</nt>?
<nt def='NT-SDDecl'>SDDecl</nt>?
<nt def="NT-S">S</nt>?
&pic;</rhs>
</prod>
<prod id='NT-VersionInfo'><lhs>VersionInfo</lhs>
<rhs><nt def="NT-S">S</nt> 'version' <nt def='NT-Eq'>Eq</nt>
(' <nt def="NT-VersionNum">VersionNum</nt> '
| " <nt def="NT-VersionNum">VersionNum</nt> ")</rhs>
</prod>
<prod id='NT-Eq'><lhs>Eq</lhs>
<rhs><nt def='NT-S'>S</nt>? '=' <nt def='NT-S'>S</nt>?</rhs></prod>
<prod id="NT-VersionNum">
<lhs>VersionNum</lhs>
<rhs>([a-zA-Z0-9_.:] | '-')+</rhs>
</prod>
<prod id='NT-Misc'><lhs>Misc</lhs>
<rhs><nt def='NT-Comment'>Comment</nt> | <nt def='NT-PI'>PI</nt> |
<nt def='NT-S'>S</nt></rhs></prod>
</prodgroup>
</scrap></p>
<p><termdef id="dt-doctype" term="Document Type Declaration">The XML
<term>document type declaration</term>
contains or points to
<termref def='dt-markupdecl'>markup declarations</termref>
that provide a grammar for a
class of documents.
This grammar is known as a document type definition,
or <term>DTD</term>.
The document type declaration can point to an external subset (a
special kind of
<termref def='dt-extent'>external entity</termref>) containing markup
declarations, or can
contain the markup declarations directly in an internal subset, or can do
both.
The DTD for a document consists of both subsets taken
together.</termdef>
</p>
<p><termdef id="dt-markupdecl" term="markup declaration">
A <term>markup declaration</term> is
an <termref def="dt-eldecl">element type declaration</termref>,
an <termref def="dt-attdecl">attribute-list declaration</termref>,
an <termref def="dt-entdecl">entity declaration</termref>, or
a <termref def="dt-notdecl">notation declaration</termref>.
</termdef>
These declarations may be contained in whole or in part
within <termref def='dt-PE'>parameter entities</termref>,
as described in the well-formedness and validity constraints below.
For fuller information, see
<specref ref="sec-physical-struct"/>.</p>
<scrap lang="ebnf" id='dtd'>
<head>Document Type Definition</head>
<prodgroup pcw2="6" pcw4="17.5" pcw5="9">
<prod id='NT-doctypedecl'><lhs>doctypedecl</lhs>
<rhs>'&lt;!DOCTYPE' <nt def='NT-S'>S</nt>
<nt def='NT-Name'>Name</nt> (<nt def='NT-S'>S</nt>
<nt def='NT-ExternalID'>ExternalID</nt>)?
<nt def='NT-S'>S</nt>? ('['
(<nt def='NT-markupdecl'>markupdecl</nt>
| <nt def='NT-PEReference'>PEReference</nt>
| <nt def='NT-S'>S</nt>)*
']'
<nt def='NT-S'>S</nt>?)? '>'</rhs>
<vc def="vc-roottype"/>
</prod>
<prod id='NT-markupdecl'><lhs>markupdecl</lhs>
<rhs><nt def='NT-elementdecl'>elementdecl</nt>
| <nt def='NT-AttlistDecl'>AttlistDecl</nt>
| <nt def='NT-EntityDecl'>EntityDecl</nt>
| <nt def='NT-NotationDecl'>NotationDecl</nt>
| <nt def='NT-PI'>PI</nt>
| <nt def='NT-Comment'>Comment</nt>
</rhs>
<vc def='vc-PEinMarkupDecl'/>
<wfc def="wfc-PEinInternalSubset"/>
</prod>
</prodgroup>
</scrap>
<p>The markup declarations may be made up in whole or in part of
the <termref def='dt-repltext'>replacement text</termref> of
<termref def='dt-PE'>parameter entities</termref>.
The productions later in this specification for
individual nonterminals (<nt def='NT-elementdecl'>elementdecl</nt>,
<nt def='NT-AttlistDecl'>AttlistDecl</nt>, and so on) describe
the declarations <emph>after</emph> all the parameter entities have been
<termref def='dt-include'>included</termref>.</p>
<vcnote id="vc-roottype">
<head>Root Element Type</head>
<p>
The <nt def='NT-Name'>Name</nt> in the document type declaration must
match the element type of the <termref def='dt-root'>root element</termref>.
</p>
</vcnote>
<vcnote id='vc-PEinMarkupDecl'>
<head>Proper Declaration/PE Nesting</head>
<p>Parameter-entity
<termref def='dt-repltext'>replacement text</termref> must be properly nested
with markup declarations.
That is to say, if either the first character
or the last character of a markup
declaration (<nt def='NT-markupdecl'>markupdecl</nt> above)
is contained in the replacement text for a
<termref def='dt-PERef'>parameter-entity reference</termref>,
both must be contained in the same replacement text.</p>
</vcnote>
<wfcnote id="wfc-PEinInternalSubset">
<head>PEs in Internal Subset</head>
<p>In the internal DTD subset,
<termref def='dt-PERef'>parameter-entity references</termref>
can occur only where markup declarations can occur, not
within markup declarations. (This does not apply to
references that occur in
external parameter entities or to the external subset.)
</p>
</wfcnote>
<p>
Like the internal subset, the external subset and
any external parameter entities referred to in the DTD
must consist of a series of complete markup declarations of the types
allowed by the non-terminal symbol
<nt def="NT-markupdecl">markupdecl</nt>, interspersed with white space
or <termref def="dt-PERef">parameter-entity references</termref>.
However, portions of the contents
of the
external subset or of external parameter entities may conditionally be ignored
by using
the <termref def="dt-cond-section">conditional section</termref>
construct; this is not allowed in the internal subset.
<scrap id="ext-Subset">
<head>External Subset</head>
<prodgroup pcw2="6" pcw4="17.5" pcw5="9">
<prod id='NT-extSubset'><lhs>extSubset</lhs>
<rhs><nt def='NT-TextDecl'>TextDecl</nt>?
<nt def='NT-extSubsetDecl'>extSubsetDecl</nt></rhs></prod>
<prod id='NT-extSubsetDecl'><lhs>extSubsetDecl</lhs>
<rhs>(
<nt def='NT-markupdecl'>markupdecl</nt>
| <nt def='NT-conditionalSect'>conditionalSect</nt>
| <nt def='NT-PEReference'>PEReference</nt>
| <nt def='NT-S'>S</nt>
)*</rhs>
</prod>
</prodgroup>
</scrap></p>
<p>The external subset and external parameter entities also differ
from the internal subset in that in them,
<termref def="dt-PERef">parameter-entity references</termref>
are permitted <emph>within</emph> markup declarations,
not only <emph>between</emph> markup declarations.</p>
<p>An example of an XML document with a document type declaration:
<eg><![CDATA[<?xml version="1.0"?>
<!DOCTYPE greeting SYSTEM "hello.dtd">
<greeting>Hello, world!</greeting>
]]></eg>
The <termref def="dt-sysid">system identifier</termref>
"<code>hello.dtd</code>" gives the URI of a DTD for the document.</p>
<p>The declarations can also be given locally, as in this
example:
<eg><![CDATA[<?xml version="1.0" encoding="UTF-8" ?>
<!DOCTYPE greeting [
<!ELEMENT greeting (#PCDATA)>
]>
<greeting>Hello, world!</greeting>
]]></eg>
If both the external and internal subsets are used, the
internal subset is considered to occur before the external subset.
<!-- 'is considered to'? boo. whazzat mean? -->
This has the effect that entity and attribute-list declarations in the
internal subset take precedence over those in the external subset.
</p>
</div2>
<div2 id='sec-rmd'>
<head>Standalone Document Declaration</head>
<p>Markup declarations can affect the content of the document,
as passed from an <termref def="dt-xml-proc">XML processor</termref>
to an application; examples are attribute defaults and entity
declarations.
The standalone document declaration,
which may appear as a component of the XML declaration, signals
whether or not there are such declarations which appear external to
the <termref def='dt-docent'>document entity</termref>.
<scrap lang="ebnf" id='fulldtd'>
<head>Standalone Document Declaration</head>
<prodgroup pcw2="4" pcw4="19.5" pcw5="9">
<prod id='NT-SDDecl'><lhs>SDDecl</lhs>
<rhs>
<nt def="NT-S">S</nt>
'standalone' <nt def='NT-Eq'>Eq</nt>
(("'" ('yes' | 'no') "'") | ('"' ('yes' | 'no') '"'))
</rhs>
<vc def='vc-check-rmd'/></prod>
</prodgroup>
</scrap></p>
<p>
In a standalone document declaration, the value "<code>yes</code>" indicates
that there
are no markup declarations external to the <termref def='dt-docent'>document
entity</termref> (either in the DTD external subset, or in an
external parameter entity referenced from the internal subset)
which affect the information passed from the XML processor to
the application.
The value "<code>no</code>" indicates that there are or may be such
external markup declarations.
Note that the standalone document declaration only
denotes the presence of external <emph>declarations</emph>; the presence, in a
document, of
references to external <emph>entities</emph>, when those entities are
internally declared,
does not change its standalone status.</p>
<p>If there are no external markup declarations, the standalone document
declaration has no meaning.
If there are external markup declarations but there is no standalone
document declaration, the value "<code>no</code>" is assumed.</p>
<p>Any XML document for which <code>standalone="no"</code> holds can
be converted algorithmically to a standalone document,
which may be desirable for some network delivery applications.</p>
<vcnote id='vc-check-rmd'>
<head>Standalone Document Declaration</head>
<p>The standalone document declaration must have
the value "<code>no</code>" if any external markup declarations
contain declarations of:</p><ulist>
<item><p>attributes with <termref def="dt-default">default</termref> values, if
elements to which
these attributes apply appear in the document without
specifications of values for these attributes, or</p></item>
<item><p>entities (other than &magicents;),
if <termref def="dt-entref">references</termref> to those
entities appear in the document, or</p>
</item>
<item><p>attributes with values subject to
<titleref href='AVNormalize'>normalization</titleref>, where the
attribute appears in the document with a value which will
change as a result of normalization, or</p>
</item>
<item>
<p>element types with <termref def="dt-elemcontent">element content</termref>,
if white space occurs
directly within any instance of those types.
</p></item>
</ulist>
</vcnote>
<p>An example XML declaration with a standalone document declaration:<eg
>&lt;?xml version="&XML.version;" standalone='yes'?></eg></p>
</div2>
<div2 id='sec-white-space'>
<head>White Space Handling</head>
<p>In editing XML documents, it is often convenient to use "white space"
(spaces, tabs, and blank lines, denoted by the nonterminal
<nt def='NT-S'>S</nt> in this specification) to
set apart the markup for greater readability. Such white space is typically
not intended for inclusion in the delivered version of the document.
On the other hand, "significant" white space that should be preserved in the
delivered version is common, for example in poetry and
source code.</p>
<p>An <termref def='dt-xml-proc'>XML processor</termref>
must always pass all characters in a document that are not
markup through to the application. A <termref def='dt-validating'>
validating XML processor</termref> must also inform the application
which of these characters constitute white space appearing
in <termref def="dt-elemcontent">element content</termref>.
</p>
<p>A special <termref def='dt-attr'>attribute</termref>
named <kw>xml:space</kw> may be attached to an element
to signal an intention that in that element,
white space should be preserved by applications.
In valid documents, this attribute, like any other, must be
<termref def="dt-attdecl">declared</termref> if it is used.
When declared, it must be given as an
<termref def='dt-enumerated'>enumerated type</termref> whose only
possible values are "<code>default</code>" and "<code>preserve</code>".
For example:<eg><![CDATA[ <!ATTLIST poem xml:space (default|preserve) 'preserve'>]]></eg></p>
<p>The value "<code>default</code>" signals that applications'
default white-space processing modes are acceptable for this element; the
value "<code>preserve</code>" indicates the intent that applications preserve
all the white space.
This declared intent is considered to apply to all elements within the content
of the element where it is specified, unless overriden with another instance
of the <kw>xml:space</kw> attribute.
</p>
<p>The <termref def='dt-root'>root element</termref> of any document
is considered to have signaled no intentions as regards application space
handling, unless it provides a value for
this attribute or the attribute is declared with a default value.
</p>
</div2>
<div2 id='sec-line-ends'>
<head>End-of-Line Handling</head>
<p>XML <termref def='dt-parsedent'>parsed entities</termref> are often stored in
computer files which, for editing convenience, are organized into lines.
These lines are typically separated by some combination of the characters
carriage-return (#xD) and line-feed (#xA).</p>
<p>To simplify the tasks of <termref def='dt-app'>applications</termref>,
wherever an external parsed entity or the literal entity value
of an internal parsed entity contains either the literal
two-character sequence "#xD#xA" or a standalone literal
#xD, an <termref def='dt-xml-proc'>XML processor</termref> must
pass to the application the single character #xA.
(This behavior can
conveniently be produced by normalizing all
line breaks to #xA on input, before parsing.)
</p>
</div2>
<div2 id='sec-lang-tag'>
<head>Language Identification</head>
<p>In document processing, it is often useful to
identify the natural or formal language
in which the content is
written.
A special <termref def="dt-attr">attribute</termref> named
<kw>xml:lang</kw> may be inserted in
documents to specify the
language used in the contents and attribute values
of any element in an XML document.
In valid documents, this attribute, like any other, must be
<termref def="dt-attdecl">declared</termref> if it is used.
The values of the attribute are language identifiers as defined
by <bibref ref="RFC1766"/>, "Tags for the Identification of Languages":
<scrap lang='ebnf'>
<head>Language Identification</head>
<prod id='NT-LanguageID'><lhs>LanguageID</lhs>
<rhs><nt def='NT-Langcode'>Langcode</nt>
('-' <nt def='NT-Subcode'>Subcode</nt>)*</rhs></prod>
<prod id='NT-Langcode'><lhs>Langcode</lhs>
<rhs><nt def='NT-ISO639Code'>ISO639Code</nt> |
<nt def='NT-IanaCode'>IanaCode</nt> |
<nt def='NT-UserCode'>UserCode</nt></rhs>
</prod>
<prod id='NT-ISO639Code'><lhs>ISO639Code</lhs>
<rhs>([a-z] | [A-Z]) ([a-z] | [A-Z])</rhs></prod>
<prod id='NT-IanaCode'><lhs>IanaCode</lhs>
<rhs>('i' | 'I') '-' ([a-z] | [A-Z])+</rhs></prod>
<prod id='NT-UserCode'><lhs>UserCode</lhs>
<rhs>('x' | 'X') '-' ([a-z] | [A-Z])+</rhs></prod>
<prod id='NT-Subcode'><lhs>Subcode</lhs>
<rhs>([a-z] | [A-Z])+</rhs></prod>
</scrap>
The <nt def='NT-Langcode'>Langcode</nt> may be any of the following:
<ulist>
<item><p>a two-letter language code as defined by
<bibref ref="ISO639"/>, "Codes
for the representation of names of languages"</p></item>
<item><p>a language identifier registered with the Internet
Assigned Numbers Authority <bibref ref='IANA'/>; these begin with the
prefix "<code>i-</code>" (or "<code>I-</code>")</p></item>
<item><p>a language identifier assigned by the user, or agreed on
between parties in private use; these must begin with the
prefix "<code>x-</code>" or "<code>X-</code>" in order to ensure that they do not conflict
with names later standardized or registered with IANA</p></item>
</ulist></p>
<p>There may be any number of <nt def='NT-Subcode'>Subcode</nt> segments; if
the first
subcode segment exists and the Subcode consists of two
letters, then it must be a country code from
<bibref ref="ISO3166"/>, "Codes
for the representation of names of countries."
If the first
subcode consists of more than two letters, it must be
a subcode for the language in question registered with IANA,
unless the <nt def='NT-Langcode'>Langcode</nt> begins with the prefix
"<code>x-</code>" or
"<code>X-</code>". </p>
<p>It is customary to give the language code in lower case, and
the country code (if any) in upper case.
Note that these values, unlike other names in XML documents,
are case insensitive.</p>
<p>For example:
<eg><![CDATA[<p xml:lang="en">The quick brown fox jumps over the lazy dog.</p>
<p xml:lang="en-GB">What colour is it?</p>
<p xml:lang="en-US">What color is it?</p>
<sp who="Faust" desc='leise' xml:lang="de">
<l>Habe nun, ach! Philosophie,</l>
<l>Juristerei, und Medizin</l>
<l>und leider auch Theologie</l>
<l>durchaus studiert mit heißem Bemüh'n.</l>
</sp>]]></eg></p>
<!--<p>The xml:lang value is considered to apply both to the contents of an
element and
(unless otherwise via attribute default values) to the
values of all of its attributes with free-text (CDATA) values. -->
<p>The intent declared with <kw>xml:lang</kw> is considered to apply to
all attributes and content of the element where it is specified,
unless overridden with an instance of <kw>xml:lang</kw>
on another element within that content.</p>
<!--
If no
value is specified for xml:lang on an element, and no default value is
defined for it in the DTD, then the xml:lang attribute of any element
takes the same value it has in the parent element, if any. The two
technical terms in the following example both have the same effective
value for xml:lang:
<p xml:lang="en">Here the keywords are
<term xml:lang="en">shift</term> and
<term>reduce</term>. ...</p>
The application, not the XML processor, is responsible for this '
inheritance' of attribute values.
-->
<p>A simple declaration for <kw>xml:lang</kw> might take
the form
<eg>xml:lang NMTOKEN #IMPLIED</eg>
but specific default values may also be given, if appropriate. In a
collection of French poems for English students, with glosses and
notes in English, the xml:lang attribute might be declared this way:
<eg><![CDATA[ <!ATTLIST poem xml:lang NMTOKEN 'fr'>
<!ATTLIST gloss xml:lang NMTOKEN 'en'>
<!ATTLIST note xml:lang NMTOKEN 'en'>]]></eg>
</p>
</div2>
</div1>
<!-- &Elements; -->
<div1 id='sec-logical-struct'>
<head>Logical Structures</head>
<p><termdef id="dt-element" term="Element">Each <termref
def="dt-xml-doc">XML document</termref> contains one or more
<term>elements</term>, the boundaries of which are
either delimited by <termref def="dt-stag">start-tags</termref>
and <termref def="dt-etag">end-tags</termref>, or, for <termref
def="dt-empty">empty</termref> elements, by an <termref
def="dt-eetag">empty-element tag</termref>. Each element has a type,
identified by name, sometimes called its "generic
identifier" (GI), and may have a set of
attribute specifications.</termdef> Each attribute specification
has a <termref
def="dt-attrname">name</termref> and a <termref
def="dt-attrval">value</termref>.
</p>
<scrap lang='ebnf'><head>Element</head>
<prod id='NT-element'><lhs>element</lhs>
<rhs><nt def='NT-EmptyElemTag'>EmptyElemTag</nt></rhs>
<rhs>| <nt def='NT-STag'>STag</nt> <nt def='NT-content'>content</nt>
<nt def='NT-ETag'>ETag</nt></rhs>
<wfc def='GIMatch'/>
<vc def='elementvalid'/>
</prod>
</scrap>
<p>This specification does not constrain the semantics, use, or (beyond
syntax) names of the element types and attributes, except that names
beginning with a match to <code>(('X'|'x')('M'|'m')('L'|'l'))</code>
are reserved for standardization in this or future versions of this
specification.
</p>
<wfcnote id='GIMatch'>
<head>Element Type Match</head>
<p>
The <nt def='NT-Name'>Name</nt> in an element's end-tag must match
the element type in
the start-tag.
</p>
</wfcnote>
<vcnote id='elementvalid'>
<head>Element Valid</head>
<p>An element is
valid if
there is a declaration matching
<nt def='NT-elementdecl'>elementdecl</nt> where the
<nt def='NT-Name'>Name</nt> matches the element type, and
one of the following holds:</p>
<olist>
<item><p>The declaration matches <kw>EMPTY</kw> and the element has no
<termref def='dt-content'>content</termref>.</p></item>
<item><p>The declaration matches <nt def='NT-children'>children</nt> and
the sequence of
<termref def="dt-parentchild">child elements</termref>
belongs to the language generated by the regular expression in
the content model, with optional white space (characters
matching the nonterminal <nt def='NT-S'>S</nt>) between each pair
of child elements.</p></item>
<item><p>The declaration matches <nt def='NT-Mixed'>Mixed</nt> and
the content consists of <termref def='dt-chardata'>character
data</termref> and <termref def='dt-parentchild'>child elements</termref>
whose types match names in the content model.</p></item>
<item><p>The declaration matches <kw>ANY</kw>, and the types
of any <termref def='dt-parentchild'>child elements</termref> have
been declared.</p></item>
</olist>
</vcnote>
<div2 id='sec-starttags'>
<head>Start-Tags, End-Tags, and Empty-Element Tags</head>
<p><termdef id="dt-stag" term="Start-Tag">The beginning of every
non-empty XML element is marked by a <term>start-tag</term>.
<scrap lang='ebnf'>
<head>Start-tag</head>
<prodgroup pcw2="6" pcw4="15" pcw5="11.5">
<prod id='NT-STag'><lhs>STag</lhs>
<rhs>'&lt;' <nt def='NT-Name'>Name</nt>
(<nt def='NT-S'>S</nt> <nt def='NT-Attribute'>Attribute</nt>)*
<nt def='NT-S'>S</nt>? '>'</rhs>
<wfc def="uniqattspec"/>
</prod>
<prod id='NT-Attribute'><lhs>Attribute</lhs>
<rhs><nt def='NT-Name'>Name</nt> <nt def='NT-Eq'>Eq</nt>
<nt def='NT-AttValue'>AttValue</nt></rhs>
<vc def='ValueType'/>
<wfc def='NoExternalRefs'/>
<wfc def='CleanAttrVals'/></prod>
</prodgroup>
</scrap>
The <nt def='NT-Name'>Name</nt> in
the start- and end-tags gives the
element's <term>type</term>.</termdef>
<termdef id="dt-attr" term="Attribute">
The <nt def='NT-Name'>Name</nt>-<nt def='NT-AttValue'>AttValue</nt> pairs are
referred to as
the <term>attribute specifications</term> of the element</termdef>,
<termdef id="dt-attrname" term="Attribute Name">with the
<nt def='NT-Name'>Name</nt> in each pair
referred to as the <term>attribute name</term></termdef> and
<termdef id="dt-attrval" term="Attribute Value">the content of the
<nt def='NT-AttValue'>AttValue</nt> (the text between the
<code>'</code> or <code>"</code> delimiters)
as the <term>attribute value</term>.</termdef>
</p>
<wfcnote id='uniqattspec'>
<head>Unique Att Spec</head>
<p>
No attribute name may appear more than once in the same start-tag
or empty-element tag.
</p>
</wfcnote>
<vcnote id='ValueType'>
<head>Attribute Value Type</head>
<p>
The attribute must have been declared; the value must be of the type
declared for it.
(For attribute types, see <specref ref='attdecls'/>.)
</p>
</vcnote>
<wfcnote id='NoExternalRefs'>
<head>No External Entity References</head>
<p>
Attribute values cannot contain direct or indirect entity references
to external entities.
</p>
</wfcnote>
<wfcnote id='CleanAttrVals'>
<head>No <code>&lt;</code> in Attribute Values</head>
<p>The <termref def='dt-repltext'>replacement text</termref> of any entity
referred to directly or indirectly in an attribute
value (other than "<code>&amp;lt;</code>") must not contain
a <code>&lt;</code>.
</p></wfcnote>
<p>An example of a start-tag:
<eg>&lt;termdef id="dt-dog" term="dog"></eg></p>
<p><termdef id="dt-etag" term="End Tag">The end of every element
that begins with a start-tag must
be marked by an <term>end-tag</term>
containing a name that echoes the element's type as given in the
start-tag:
<scrap lang='ebnf'>
<head>End-tag</head>
<prodgroup pcw2="6" pcw4="15" pcw5="11.5">
<prod id='NT-ETag'><lhs>ETag</lhs>
<rhs>'&lt;/' <nt def='NT-Name'>Name</nt>
<nt def='NT-S'>S</nt>? '>'</rhs></prod>
</prodgroup>
</scrap>
</termdef></p>
<p>An example of an end-tag:<eg>&lt;/termdef></eg></p>
<p><termdef id="dt-content" term="Content">The
<termref def='dt-text'>text</termref> between the start-tag and
end-tag is called the element's
<term>content</term>:
<scrap lang='ebnf'>
<head>Content of Elements</head>
<prodgro