Interoperability:

The number of literals for each value has been kept small; for manydatatypes there is a one-to-one mapping between literals and values.This makes it easy to exchange the values between different systems.In many cases, conversion from locale-dependent representations willbe required on both the originator and the recipient side, both forcomputer processing and for interaction with humans.

Basic readability:

Textual, rather than binary, literals are used.This makes hand editing, debugging, and similar activities possible.

Ease of parsing and serializing:

Where possible, literals correspond to those found in commonprogramming languages and libraries.

2.3.1 Canonical Lexical Representation

While the datatypes defined in this specification have, for the most part,a single lexical representation i.e. each value in the datatype's·value space· is denoted by a single literal in its·lexical space·, this is not always the case. Theexample in the previous section showed two literals for the datatypefloat which denote the same value. Similarly, there·may· beseveral literals for one of the date or time datatypes that denote thesame value using different timezone indicators.

[Definition:]  Acanonical lexical representationis a set of literals from among the valid set of literalsfor a datatype such that there is a one-to-one mapping between literalsin thecanonical lexical representation andvalues in the·value space·.

2.4.1 Fundamental facets

[Definition:]  Afundamental facet is an abstract property whichserves to semantically characterize the values in a·value space·.

Allfundamental facets are fully described inFundamental Facets (§4.2).

2.4.2 Constraining or Non-fundamental facets

[Definition:]  Aconstraining facet is an optional property that can beapplied to a datatype to constrain its·value space·.

Constraining the·value space· consequently constrainsthe·lexical space·. Adding·constraining facet·s to a·base type·is described inDerivation by restriction (§4.1.2.1).

Allconstraining facets are fully described inConstraining Facets (§4.3).

2.5.1 Atomic vs. list vs. union datatypes

The first distinction to be made is that between·atomic·,·list· and·union·datatypes.

• [Definition:]  Atomic datatypesare those having values which are regarded by this specification asbeing indivisible.
• [Definition:]  Listdatatypes are those having values each of which consists of afinite-length (possibly empty) sequence of values of an·atomic· datatype.
• [Definition:]  Uniondatatypes are those whose·value space·s and·lexical space·s are the union ofthe·value space·s and·lexical space·s of one or more other datatypes.

For example, a single token which·match·esNmtoken from[XML 1.0 (Second Edition)] could be the value of an·atomic·datatype (NMTOKEN); while a sequence of such tokenscould be the value of a·list· datatype(NMTOKENS).

<simpleType name='sizes'>  <list itemType='decimal'/></simpleType>

<cerealSizes xsi:type='sizes'> 8 10.5 12 </cerealSizes>

<simpleType name='listOfString'>  <list itemType='string'/></simpleType>

<someElement xsi:type='listOfString'>this is not list item 1this is not list item 2this is not list item 3</someElement>

<xs:simpleType name='myList'><xs:list itemType='xs:integer'/></xs:simpleType><xs:simpleType name='myRestrictedList'><xs:restriction base='myList'><xs:pattern value='123 (\d+\s)*456'/></xs:restriction></xs:simpleType><someElement xsi:type='myRestrictedList'>123 456</someElement><someElement xsi:type='myRestrictedList'>123 987 456</someElement><someElement xsi:type='myRestrictedList'>123 987 567 456</someElement>

  <attributeGroup name="occurs">    <attribute name="minOccurs" type="nonNegativeInteger"    use="optional" default="1"/>    <attribute name="maxOccurs"use="optional" default="1">      <simpleType>        <union>          <simpleType>            <restriction base='nonNegativeInteger'/>          </simpleType>          <simpleType>            <restriction base='string'>              <enumeration value='unbounded'/>            </restriction>          </simpleType>        </union>      </simpleType>    </attribute>  </attributeGroup>

  <xsd:element name='size'>    <xsd:simpleType>      <xsd:union>        <xsd:simpleType>          <xsd:restriction base='integer'/>        </xsd:simpleType>        <xsd:simpleType>          <xsd:restriction base='string'/>        </xsd:simpleType>      </xsd:union>    </xsd:simpleType>  </xsd:element>

  <size>1</size>  <size>large</size>  <size xsi:type='xsd:string'>1</size>

2.5.2 Primitive vs. derived datatypes

Next, we distinguish between·primitive· and·derived· datatypes.

• [Definition:]  Primitivedatatypes are those that are not defined in terms of other datatypes;they existab initio.
• [Definition:]  Deriveddatatypes are those that are defined in terms of other datatypes.

For example, in this specification,float is a well-definedmathematicalconcept that cannot be defined in terms of other datatypes, whileainteger is a special case of the more general datatypedecimal.

[Definition:]  The simple ur-type definition is a special restriction of theur-type definitionwhose name isanySimpleType in the XML Schema namespace.anySimpleType can beconsidered as the·base type· of all·primitive·datatypes.anySimpleType is considered to have an unconstrained lexical space and a·value space· consisting of the union of the·value space·s of all the·primitive·datatypes and the set of all lists of all members of the·value space·s of all the·primitive· datatypes.

The datatypes defined by this specification fall into boththe·primitive· and·derived·categories. It is felt that a judiciously chosen set of·primitive· datatypes will serve the widestpossible audience by providing a set of convenient datatypes thatcan be used as is, as well as providing a rich enough base fromwhich the variety of datatypes needed by schema designers can be·derived·.

In the example above,integer is·derived·fromdecimal.

As described in more detail inXML Representation of Simple Type Definition Schema Components (§4.1.2),each·user-derived· datatype·must·be defined in terms of another datatype in one of three ways: 1) by assigning·constraining facet·s which serve torestrict the·value space· of the·user-derived·datatype to a subset of that of the·base type·; 2) by creatinga·list· datatype whose·value space·consists of finite-length sequences of values of its·itemType·; or 3) by creating a·union·datatype whose·value space· consists of the union of the·value space·s of its·memberTypes·.

2.5.3 Built-in vs. user-derived datatypes

• [Definition:]  Built-indatatypes are those which are defined in this specification,and can be either·primitive· or·derived·;
• [Definition:]  User-derived datatypes are those·derived·datatypes that are defined by individual schema designers.

Conceptually there is no difference between the·built-in· ·derived· datatypesincluded in this specification and the·user-derived·datatypes which will be created by individual schema designers.The·built-in· ·derived· datatypesare those which are believed to be so common that if they were notdefined in this specification many schema designers would end up"reinventing" them. Furthermore, including these·derived· datatypes in this specification serves todemonstrate the mechanics and utility of the datatype generationfacilities of this specification.

3.2.1 string

[Definition:]  Thestring datatyperepresents character strings in XML. The·value space·ofstring is the set of finite-length sequences ofcharacters (as defined in[XML 1.0 (Second Edition)]) that·match· theChar production from[XML 1.0 (Second Edition)].Acharacter is an atomic unit ofcommunication; it is not further specified except to note that everycharacter has a correspondingUniversal Character Set code point, which is an integer.

3.2.2 boolean

[Definition:]  boolean has the·value space· required to support the mathematicalconcept of binary-valued logic: {true, false}.

3.2.3 decimal

[Definition:]  decimalrepresents a subset of the real numbers, which can be represented by decimal numerals.The·value space· ofdecimalis the set of numbers that can be obtained by multiplying an integer by a non-positivepower of ten, i.e., expressible asi × 10^-nwherei andn are integersandn >= 0.Precision is not reflected in this value space;the number 2.0 is not distinct from the number 2.00.The·order-relation· ondecimalis the order relation on real numbers, restrictedto this subset.

3.2.4 float

[Definition:]  floatis patterned after the IEEE single-precision 32-bit floating point type[IEEE 754-1985]. The basic·value space· offloat consists of the valuesm × 2^e, wheremis an integer whose absolute value is less than2^24, ande is an integerbetween -149 and 104, inclusive. In addition to the basic·value space· described above, the·value space· offloat also contains thefollowingthreespecial values:positive and negative infinity and not-a-number(NaN).The·order-relation· onfloatis:x < y iff y - x is positivefor x and y in the value space.Positive infinity is greater than all other non-NaN values.NaN equals itself but is·incomparable· with (neither greater than nor less than)any other value in the·value space·.

A literal in the·lexical space· representing adecimal numberd maps to the normalized valuein the·value space· offloat that isclosest tod in the sense defined by[Clinger, WD (1990)]; ifd isexactly halfway between two such values then the even value is chosen.

3.2.5 double

[Definition:]  Thedoubledatatypeis patterned after theIEEE double-precision 64-bit floating pointtype[IEEE 754-1985]. The basic·value space·ofdouble consists of the valuesm × 2^e, wheremis an integer whose absolute value is less than2^53, ande is aninteger between -1075 and 970, inclusive. In addition to the basic·value space· described above, the·value space· ofdouble also containsthe followingthreespecial values:positive and negative infinity and not-a-number(NaN).The·order-relation· ondoubleis:x < y iff y - x is positivefor x and y in the value space.Positive infinity is greater than all other non-NaN values.NaN equals itself but is·incomparable· with (neither greater than nor less than)any other value in the·value space·.

A literal in the·lexical space· representing adecimal numberd maps to the normalized valuein the·value space· ofdouble that isclosest tod; ifd isexactly halfway between two such values then the even value is chosen.This is thebest approximation ofd([Clinger, WD (1990)],[Gay, DM (1990)]), which is moreaccurate than the mapping required by[IEEE 754-1985].

3.2.6 duration

[Definition:]  duration represents a duration of time.The·value space· ofduration isa six-dimensional space where the coordinatesdesignate the Gregorian year, month, day, hour, minute, and second components defined in§ 5.5.3.2 of[ISO 8601],respectively. These components are orderedin their significance by their order of appearance i.e. as year, month, day,hour, minute, and second.

<simpleType name='SQL-Year-Month-Interval'>    <restriction base='duration'>      <pattern value='P\p{Nd}{4}Y\p{Nd}{2}M'/>    </restriction></simpleType>

3.2.7 dateTime

[Definition:]  dateTime values may be viewed as objects with integer-valuedyear, month, day, hour and minute properties, a decimal-valued second property,and a boolean timezoned property.Each such object also has one decimal-valuedmethod or computed property, timeOnTimeline, whose value is always a decimalnumber; the values are dimensioned in seconds, the integer 0 is0001-01-01T00:00:00 and the value of timeOnTimeline for otherdateTimevalues is computed using the Gregorian algorithm as modified for leap-seconds.The timeOnTimeline values form two related "timelines", one for timezonedvalues and one for non-timezoned values.Each timeline is a copy of the·value space· ofdecimal,with integers given units of seconds.

The·value space· ofdateTime is closely related to the dates and times described in ISO 8601.For clarity, the text above specifies a particular origin point for thetimeline.It should be noted, however, that schema processors need not expose thetimeOnTimeline value to schema users, and there is no requirement that atimeline-based implementation use the particular origin described here inits internal representation.Other interpretations of the·value space· which lead to thesame results (i.e., are isomorphic) are of course acceptable.

All timezoned times are Coordinated Universal Time (UTC, sometimes called"Greenwich Mean Time"). Other timezones indicated in lexical representationsare converted to UTC during conversion of literals to values."Local" or untimezoned times are presumed to be the time in the timezone of someunspecified locality as prescribed by the appropriate legal authority;currently there are no legally prescribed timezones which are durationswhose magnitude is greater than 14 hours. The value of each numeric-valued property(other than timeOnTimeline) is limited to the maximum value within the intervaldetermined by the next-higher property. For example, the day value can never be 32,and cannot even be 29 for month 02 and year 2002 (February 2002).

3.2.8 time

[Definition:]  timerepresents an instant of time that recurs every day. The·value space· oftime is the spaceoftime of day values as defined in § 5.3 of[ISO 8601]. Specifically, it is a set of zero-duration dailytime instances.

Since the lexical representation allows an optional time zoneindicator,time values are partially ordered because it maynot be able to determine the order of two values one of which has atime zone and the other does not. The order relation ontime values is theOrder relation on dateTime (§3.2.7.4) using an arbitrary date. See alsoAdding durations to dateTimes (§E). Pairs oftime values with or without time zone indicators are totally ordered.

3.2.9 date

[Definition:]  The·value space· ofdateconsists of top-open intervals of exactly one day in length on the timelines ofdateTime, beginning on the beginning moment of each day (ineach timezone), i.e. '00:00:00', up to but not including '24:00:00' (which isidentical with '00:00:00' of the next day). For nontimezoned values, the top-openintervals disjointly cover the nontimezoned timeline, one per day. For timezonedvalues, the intervals begin at every minute and therefore overlap.

A "date object" is an object with year, month, and day properties just like thoseofdateTime objects, plus an optionaltimezone-valuedtimezone property. (As with values ofdateTime timezones are aspecial case of durations.)Just as adateTime object corresponds to a point on one of thetimelines, adate object corresponds to an interval on oneof the two timelines as just described.

Timezoneddate values track the starting moment of their day, asdetermined by their timezone; said timezone is generally recoverable forcanonical representations.[Definition:]  Therecoverable timezone is that duration whichis the result of subtracting the first moment (or any moment) of the timezoneddate from the first moment (or the corresponding moment) UTC on thesamedate.·recoverable timezone·s arealways durations between '+12:00' and '-11:59'. This "timezone normalization"(which follows automatically from the definition of thedate·value space·) is explained more inLexical representation (§3.2.9.1).

For example: the first moment of 2002-10-10+13:00 is 2002-10-10T00:00:00+13,which is 2002-10-09T11:00:00Z, which is also the first moment of 2002-10-09-11:00.Therefore 2002-10-10+13:00 is 2002-10-09-11:00;they are the same interval.

3.2.10 gYearMonth

[Definition:]  gYearMonth represents aspecific gregorian month in a specific gregorian year. The·value space· ofgYearMonthis the set of Gregorian calendar months as defined in § 5.2.1 of[ISO 8601]. Specifically, it is a set of one-month long,non-periodic instancese.g. 1999-10 to represent the whole month of 1999-10, independent ofhow many days this month has.

Since the lexical representation allows an optional time zoneindicator,gYearMonth values are partially ordered because it maynot be possible to unequivocally determine the order of two values one ofwhich has a time zone and the other does not. IfgYearMonthvalues are considered as periods of time, the order relation ongYearMonth values is the order relation on their starting instants.This is discussed inOrder relation on dateTime (§3.2.7.4). See alsoAdding durations to dateTimes (§E). Pairs ofgYearMonthvalues with or without time zone indicators are totally ordered.

3.2.11 gYear

[Definition:]  gYear represents agregorian calendar year. The·value space· ofgYear is the set of Gregorian calendar years as defined in§ 5.2.1 of[ISO 8601]. Specifically, it is a set of one-yearlong, non-periodic instancese.g. lexical 1999 to represent the whole year 1999, independent ofhow many months and days this year has.

Since the lexical representation allows an optional time zoneindicator,gYear values are partially ordered because it maynot be possible to unequivocally determine the order of two values one of which has atime zone and the other does not. IfgYear values are considered as periods of time, the order relationongYear values is the order relation on their starting instants.This is discussed inOrder relation on dateTime (§3.2.7.4). See alsoAdding durations to dateTimes (§E). Pairs ofgYear values with or without time zone indicators are totally ordered.

3.2.12 gMonthDay

[Definition:]  gMonthDay is a gregorian date that recurs, specifically a day ofthe year such as the third of May. Arbitrary recurring dates are notsupported by this datatype. The·value space· ofgMonthDay is the set ofcalendardates, as defined in § 3 of[ISO 8601]. Specifically,it is a set of one-day long, annually periodic instances.

Since the lexical representation allows an optional time zoneindicator,gMonthDay values are partially ordered because it maynot be possible to unequivocally determine the order of two values one of which has atime zone and the other does not. IfgMonthDay values are considered as periods of time,in an arbitrary leap year, the order relationongMonthDay values is the order relation on their starting instants.This is discussed inOrder relation on dateTime (§3.2.7.4). See alsoAdding durations to dateTimes (§E). Pairs ofgMonthDay values with or without time zone indicators are totally ordered.

3.2.13 gDay

[Definition:]  gDay is a gregorian day that recurs, specifically a dayof the month such as the 5th of the month. Arbitrary recurring daysare not supported by this datatype. The·value space·ofgDay is the space of a set ofcalendardates as defined in § 3 of[ISO 8601]. Specifically,it is a set of one-day long, monthly periodic instances.

This datatype can be used to represent a specific day of the month.To say, for example, that I get my paycheck on the 15th of each month.

Since the lexical representation allows an optional time zoneindicator,gDay values are partially ordered because it maynot be possible to unequivocally determine the order of two values one ofwhich has a time zone and the other does not. IfgDay values are considered as periods of time,in an arbitrary month that has 31 days,the order relationongDay values is the order relation on their starting instants.This is discussed inOrder relation on dateTime (§3.2.7.4). See alsoAdding durations to dateTimes (§E). Pairs ofgDayvalues with or without time zone indicators are totally ordered.

3.2.14 gMonth

[Definition:]  gMonth is a gregorian month that recurs every year.The·value space·ofgMonth is the space of a set ofcalendarmonths as defined in § 3 of[ISO 8601]. Specifically,it is a set of one-month long, yearly periodic instances.

This datatype can be used to represent a specific month.To say, for example, that Thanksgiving falls in the month of November.

Since the lexical representation allows an optional time zoneindicator,gMonth values are partially ordered because it maynot be possible to unequivocally determine the order of two values one of which has atime zone and the other does not. IfgMonth values are considered as periods of time, the order relationongMonth is the order relation on their starting instants.This is discussed inOrder relation on dateTime (§3.2.7.4). See alsoAdding durations to dateTimes (§E). Pairs ofgMonthvalues with or without time zone indicators are totally ordered.

3.2.15 hexBinary

[Definition:]  hexBinary representsarbitrary hex-encoded binary data. The·value space· ofhexBinary is the set of finite-length sequences of binaryoctets.

3.2.16 base64Binary

[Definition:]  base64Binaryrepresents Base64-encoded arbitrary binary data. The·value space· ofbase64Binary is the set of finite-length sequences of binaryoctets. Forbase64Binary data theentire binary stream is encoded using the Base64Alphabet in[RFC 2045].

The lexical forms ofbase64Binary values are limited to the 65 charactersof the Base64 Alphabet defined in[RFC 2045], i.e.,a-z,A-Z,0-9, the plus sign (+), the forward slash (/) and theequal sign (=), together with the characters defined in[XML 1.0 (Second Edition)] as white space.No other characters are allowed.

For compatibility with older mail gateways,[RFC 2045] suggests thatbase64 data should have lines limited to at most 76 characters in length. Thisline-length limitation is not mandated in the lexical forms ofbase64Binarydata and must not be enforced by XML Schema processors.

The lexical space ofbase64Binary is given by the following grammar(the notation is that used in[XML 1.0 (Second Edition)]); legal lexical forms must matchtheBase64Binary production.

Base64Binary  ::=  ((B64S B64S B64S B64S)*
                     ((B64S B64S B64S B64) |
                      (B64S B64S B16S '=') |
                      (B64S B04S '=' #x20? '=')))?

B64S         ::= B64 #x20?

B16S         ::= B16 #x20?

B04S         ::= B04 #x20?

B04         ::=  [AQgw]
B16         ::=  [AEIMQUYcgkosw048]
B64         ::=  [A-Za-z0-9+/]

Note that this grammar requires the number of non-whitespace characters in the lexicalform to be a multiple of four, and for equals signs to appear only at the end of thelexical form; strings which do not meet these constraints are not legal lexical formsofbase64Binary because they cannot successfully be decoded by base64decoders.

The canonical lexical form of abase64Binary data value is the base64encoding of the value which matches the Canonical-base64Binary production in the followinggrammar:

Canonical-base64Binary  ::=  (B64B64 B64 B64)*
                               ((B64 B64 B16 '=') | (B64 B04 '=='))?

The length of abase64Binary value is the number of octets it contains.This may be calculated from the lexical form by removing whitespace and padding charactersand performing the calculation shown in the pseudo-code below:

lex2    := killwhitespace(lexform)    -- remove whitespace characters
lex3    := strip_equals(lex2)         -- strip padding characters at end
length  := floor (length(lex3) * 3 / 4)         -- calculate length

Note on encoding:[RFC 2045] explicitly references US-ASCII encoding. However,decoding ofbase64Binary data in an XML entity is to be performed on theUnicode characters obtained after character encoding processing as specified by[XML 1.0 (Second Edition)]

3.2.17 anyURI

[Definition:]  anyURI represents a Uniform Resource Identifier Reference(URI). AnanyURI value can be absolute or relative, and mayhave an optional fragment identifier (i.e., it may be a URI Reference). Thistype should be used to specify the intention that the value fulfillsthe role of a URI as defined by[RFC 2396], as amended by[RFC 2732].

The mapping fromanyURI values to URIs is asdefined by the URI reference escaping proceduredefined inSection 5.4Locator Attributeof[XML Linking Language] (see also Section 8Character Encoding in URI Referencesof[Character Model]). This meansthat a wide range of internationalized resource identifiers can be specifiedwhen ananyURI is called for, and still be understood asURIs per[RFC 2396], as amended by[RFC 2732],where appropriate to identify resources.

3.2.18 QName

[Definition:]  QName representsXML qualified names.The·value space· ofQName is the set oftuples {namespace name,local part},wherenamespace nameis ananyURIandlocal part isanNCName.The·lexical space· ofQName is the setof strings that·match· theQName production of[Namespaces in XML].

3.2.19 NOTATION

[Definition:]  NOTATIONrepresents theNOTATION attributetype from[XML 1.0 (Second Edition)]. The·value space·ofNOTATION is the set ofQNamesof notations declared in the current schema.The·lexical space· ofNOTATION is the setof all names ofnotationsdeclared in the current schema (in the form ofQNames).

For compatibility (seeTerminology (§1.4))NOTATIONshould be used only on attributesand should only be used in schemas with notarget namespace.

3.3.1 normalizedString

[Definition:]  normalizedStringrepresents white space normalized strings.The·value space· ofnormalizedString is theset of strings that do notcontain the carriage return (#xD), line feed (#xA) nor tab (#x9) characters.The·lexical space· ofnormalizedString is theset of strings that do notcontain the carriage return (#xD),line feed (#xA)nor tab (#x9) characters.The·base type· ofnormalizedString isstring.

3.3.2 token

[Definition:]  tokenrepresents tokenized strings.The·value space· oftoken is theset of strings that do notcontain thecarriage return (#xD),line feed (#xA) nor tab (#x9) characters, that have noleading or trailing spaces (#x20) and that have no internal sequencesof two or more spaces.The·lexical space· oftoken is theset of strings that do not contain thecarriage return (#xD),line feed (#xA) nor tab (#x9) characters, that have noleading or trailing spaces (#x20) and that have no internal sequencesof two or more spaces.The·base type· oftoken isnormalizedString.

3.3.3 language

[Definition:]  languagerepresents natural language identifiers as defined byby[RFC 3066].The·value space· oflanguage is theset of all strings that are valid language identifiers as defined[RFC 3066].The·lexical space· oflanguage is the set of all strings thatconform to the pattern[a-zA-Z]{1,8}(-[a-zA-Z0-9]{1,8})*.The·base type· oflanguage istoken.

3.3.4 NMTOKEN

[Definition:]  NMTOKEN representstheNMTOKEN attribute typefrom[XML 1.0 (Second Edition)]. The·value space· ofNMTOKEN is the set of tokens that·match·theNmtoken production in[XML 1.0 (Second Edition)]. The·lexical space· ofNMTOKEN is the set of strings that·match·theNmtoken production in[XML 1.0 (Second Edition)]. The·base type· ofNMTOKEN istoken.

For compatibility (seeTerminology (§1.4))NMTOKENshould be used only on attributes.

3.3.5 NMTOKENS

[Definition:]  NMTOKENSrepresents theNMTOKENS attributetype from[XML 1.0 (Second Edition)]. The·value space·ofNMTOKENS is the set of finite, non-zero-length sequences of·NMTOKEN·s. The·lexical space·ofNMTOKENS is the set of space-separated lists of tokens,of which each token is in the·lexical space· ofNMTOKEN. The·itemType· ofNMTOKENS isNMTOKEN.

For compatibility (seeTerminology (§1.4))NMTOKENS should be used only on attributes.

3.3.6 Name

[Definition:]  NamerepresentsXML Names.The·value space· ofName isthe set of all strings which·match· theName production of[XML 1.0 (Second Edition)]. The·lexical space· ofName is the set of all strings which·match·theName production of[XML 1.0 (Second Edition)]. The·base type· ofNameistoken.

3.3.7 NCName

[Definition:]  NCName represents XML"non-colonized" Names. The·value space· ofNCName is the set of all strings which·match·theNCName production of[Namespaces in XML]. The·lexical space· ofNCName is the set of all strings which·match·theNCName production of[Namespaces in XML]. The·base type· ofNCName isName.

3.3.8 ID

[Definition:]  ID represents theID attribute type from[XML 1.0 (Second Edition)]. The·value space· ofID is the set of all strings that·match·theNCName production in[Namespaces in XML]. The·lexical space· ofID is the set of allstrings that·match· theNCName production in[Namespaces in XML].The·base type· ofID isNCName.

For compatibility (seeTerminology (§1.4))ID should be used only on attributes.

3.3.9 IDREF

[Definition:]  IDREF represents theIDREF attribute type from[XML 1.0 (Second Edition)]. The·value space· ofIDREF is the set of all strings that·match·theNCName production in[Namespaces in XML]. The·lexical space· ofIDREF is the set ofstrings that·match· theNCName production in[Namespaces in XML].The·base type· ofIDREF isNCName.

For compatibility (seeTerminology (§1.4)) this datatypeshould be used only on attributes.

3.3.10 IDREFS

[Definition:]  IDREFS represents theIDREFS attribute type from[XML 1.0 (Second Edition)]. The·value space· ofIDREFS is the set of finite, non-zero-length sequences ofIDREFs.The·lexical space· ofIDREFS is theset of space-separated lists of tokens, of which each token is in the·lexical space· ofIDREF.The·itemType· ofIDREFS isIDREF.

For compatibility (seeTerminology (§1.4))IDREFSshould be used only on attributes.

3.3.11 ENTITY

[Definition:]  ENTITY represents theENTITY attribute type from[XML 1.0 (Second Edition)]. The·value space· ofENTITY is the set of all strings that·match·theNCName production in[Namespaces in XML] and have been declared as anunparsed entity inadocument type definition.The·lexical space· ofENTITY is the setof all strings that·match· theNCName production in[Namespaces in XML].The·base type· ofENTITY isNCName.

For compatibility (seeTerminology (§1.4))ENTITYshould be used only on attributes.

3.3.12 ENTITIES

[Definition:]  ENTITIESrepresents theENTITIES attributetype from[XML 1.0 (Second Edition)]. The·value space·ofENTITIES is the set of finite, non-zero-length sequences of·ENTITY·s that have been declared asunparsed entitiesin adocument type definition.The·lexical space· ofENTITIES is theset of space-separated lists of tokens, of which each token is in the·lexical space· ofENTITY.The·itemType· ofENTITIES isENTITY.

For compatibility (seeTerminology (§1.4))ENTITIESshould be used only on attributes.

3.3.13 integer

[Definition:]  integer is·derived· fromdecimal by fixing thevalue of·fractionDigits· to be 0anddisallowing the trailing decimal point.This results in the standardmathematical concept of the integer numbers. The·value space· ofinteger is the infiniteset {...,-2,-1,0,1,2,...}. The·base type· ofinteger isdecimal.

3.3.14 nonPositiveInteger

[Definition:]  nonPositiveInteger is·derived· frominteger by setting the value of·maxInclusive· to be 0. This results in thestandard mathematical concept of the non-positive integers.The·value space· ofnonPositiveIntegeris the infinite set {...,-2,-1,0}. The·base type·ofnonPositiveInteger isinteger.

3.3.15 negativeInteger

[Definition:]  negativeInteger is·derived· fromnonPositiveInteger by setting the value of·maxInclusive· to be -1. This results in thestandard mathematical concept of the negative integers. The·value space· ofnegativeIntegeris the infinite set {...,-2,-1}. The·base type·ofnegativeInteger isnonPositiveInteger.

3.3.16 long

[Definition:]  long is·derived· frominteger by setting thevalue of·maxInclusive· to be 9223372036854775807and·minInclusive· to be -9223372036854775808.The·base type· oflong isinteger.

3.3.17 int

[Definition:]  intis·derived· fromlong by setting thevalue of·maxInclusive· to be 2147483647 and·minInclusive· to be -2147483648. The·base type· ofint islong.

3.3.18 short

[Definition:]  short is·derived· fromint by setting thevalue of·maxInclusive· to be 32767 and·minInclusive· to be -32768. The·base type· ofshort isint.

3.3.19 byte

[Definition:]  byteis·derived· fromshortby setting the value of·maxInclusive· to be 127and·minInclusive· to be -128.The·base type· ofbyte isshort.

3.3.20 nonNegativeInteger

[Definition:]  nonNegativeInteger is·derived· frominteger by setting the value of·minInclusive· to be 0. This results in thestandard mathematical concept of the non-negative integers. The·value space· ofnonNegativeIntegeris the infinite set {0,1,2,...}. The·base type· ofnonNegativeInteger isinteger.

3.3.21 unsignedLong

[Definition:]  unsignedLong is·derived· fromnonNegativeInteger by setting the value of·maxInclusive· to be 18446744073709551615.The·base type· ofunsignedLong isnonNegativeInteger.

3.3.22 unsignedInt

[Definition:]  unsignedInt is·derived· fromunsignedLong by setting the value of·maxInclusive· to be 4294967295. The·base type· ofunsignedInt isunsignedLong.

3.3.23 unsignedShort

[Definition:]  unsignedShort is·derived· fromunsignedInt by setting the value of·maxInclusive· to be 65535. The·base type· ofunsignedShort isunsignedInt.

3.3.24 unsignedByte

[Definition:]  unsignedByte is·derived· fromunsignedShort by setting the value of·maxInclusive· to be 255. The·base type· ofunsignedByte isunsignedShort.

3.3.25 positiveInteger

[Definition:]  positiveInteger is·derived· fromnonNegativeInteger by setting the value of·minInclusive· to be 1. This results in the standardmathematical concept of the positive integer numbers.The·value space· ofpositiveIntegeris the infinite set {1,2,...}. The·base type· ofpositiveInteger isnonNegativeInteger.

4.1.1 The Simple Type Definition Schema Component

The Simple Type Definition schema component has the following properties:

Datatypes are identified by their{name}and{target namespace}. Exceptfor anonymous datatypes (those with no{name}),datatype definitions·must· be uniquely identifiedwithin a schema.

If{variety} is·atomic·then the·value space· of the datatype defined willbe a subset of the·value space· of{base type definition} (which is a subset of the·value space· of{primitive type definition}).If{variety} is·list·then the·value space· of the datatype defined willbe the set of finite-length sequence of values from the·value space· of{item type definition}.If{variety} is·union· then the·value space· of the datatype defined will be theunion of the·value space·s of each datatype in{member type definitions}.

If{variety} is·atomic·then the{variety} of{base type definition}must be·atomic·.If{variety} is·list·then the{variety} of{item type definition}must be either·atomic· or·union·.If{variety} is·union·then{member type definitions} must be a list of datatype definitions.

The value of{facets} consists of the set of·facet·s specified directly in the datatype definitionunioned with the possibly empty set of{facets} of{base type definition}.

The value of{fundamental facets} consists of the set of·fundamental facet·s and their values.

If{final} is the empty set then the type can be usedin deriving other types; the explicit valuesrestriction,list andunion prevent further derivationsby·restriction·,·list· and·union· respectively.

4.1.2 XML Representation of Simple Type Definition Schema Components

The XML representation for aSimple Type Definition schema componentis a<simpleType> element information item. Thecorrespondences between the properties of the information item andproperties of the component are as follows:

A·derived· datatype can be·derived·from a·primitive· datatype or another·derived· datatype by one of three means:byrestriction, bylist or byunion.

<simpleType name='Sku'>    <restriction base='string'>      <pattern value='\d{3}-[A-Z]{2}'/>    </restriction></simpleType>

<simpleType name='listOfFloat'>  <list itemType='float'/></simpleType>

<xsd:attribute name="size">  <xsd:simpleType>    <xsd:union>      <xsd:simpleType>        <xsd:restriction base="xsd:positiveInteger">          <xsd:minInclusive value="8"/>          <xsd:maxInclusive value="72"/>        </xsd:restriction>      </xsd:simpleType>      <xsd:simpleType>        <xsd:restriction base="xsd:NMTOKEN">          <xsd:enumeration value="small"/>          <xsd:enumeration value="medium"/>          <xsd:enumeration value="large"/>        </xsd:restriction>      </xsd:simpleType>    </xsd:union>  </xsd:simpleType></xsd:attribute>

<p><font size='large'>A header</font></p><p><font size='12'>this is a test</font></p>

4.1.3 Constraints on XML Representation of Simple Type Definition

4.1.4 Simple Type Definition Validation Rules

4.1.5 Constraints on Simple Type Definition Schema Components

4.1.6 Simple Type Definition for anySimpleType

There is a simple type definition nearly equivalent to the simple versionof theur-type definition presentin every schema by definition. It has the following properties:

4.2.1 equal

Every·value space· supports the notion of equality,with the following rules:

• for anya andb inthe·value space·,eithera is equal tob,denoteda = b, orais not equal tob, denoteda != b
• there is no paira andbfrom the·value space· such that botha = b anda != b
• for alla in the·value space·,a = a
• for anya andbin the·value space·,a = b if and only ifb = a
• for anya,b andc in the·value space·,ifa = b andb = c, thena = c
• for anya andbin the·value space·ifa = b, thenaandb cannot be distinguished(i.e., equality is identity)
• the·value space·s of all·primitive· datatypes are disjoint (they do notshare any values)

On every datatype, the operation Equal is defined in terms of the equalityproperty of the·value space·: for any valuesa, b drawn from the·value space·,Equal(a,b) istrue ifa = b, and false otherwise.

Note that in consequence of the above:

• given·value space· A and·value space· B whereA andB are disjoint,every pair of valuesa fromAandb fromB,a != b
• two values which are members of the·value space·of the same·primitive· datatype may always becompared with each other
• if a datatypeT is·derived· by·union· from·memberTypes· A, B, ...then the·value space· ofT is theunion of·value space·s of its·memberTypes· A, B, ....Some values in the·value space· ofT are also values in the·value space· ofA.Other values in the·value space· ofT will be values in the·value space· ofB and so on.Values in the·value space· ofTwhich are also in the·value space· ofA can be compared with other values in the·value space· ofA accordingto the above rules. Similarly for values of typeT andB and all the other·memberTypes·.
• if a datatypeT' is·derived·by·restriction· from an atomic datatypeTthen the·value space· ofT' isa subset of the·value space· ofT.Values in the·value space·s ofT andT' can be comparedaccording to the above rules
• if datatypesT' andT'' are·derived· by·restriction· from acommon atomic ancestorT then the·value space·s ofT' andT'' may overlap. Values in the·value space·sofT' andT'' can becompared according to the above rules

4.2.2 ordered

[Definition:]  Anorder relation on a·value space·is a mathematical relation that imposes a·total order· or a·partial order· on themembers of the·value space·.

[Definition:]  A·value space·, and hence a datatype, is said to beordered if there exists an·order-relation· defined for that·value space·.

[Definition:]  Apartial order is an·order-relation·that isirreflexive,asymmetric andtransitive.

A·partial order· has the following properties:

• for noa in the·value space·,a < a(irreflexivity)
• for alla andbin the·value space·,a < bimplies not(b < a)(asymmetry)
• for alla,bandc in the·value space·,a < b andb < cimpliesa < c(transitivity)

The notationa <> b is used to indicate thecase whena != b and neithera < b norb < a.For any valuesa andbfrom different·primitive· ·value space·s,a <> b.

[Definition:]  Whena <> b,a andb areincomparable,[Definition:]  otherwise they arecomparable.

[Definition:]  Atotal order is an·partial order·such that for noa andbis it the case thata <> b.

A·total order· has all of the properties specifiedabove for·partial order·, plusthe following property:

• for alla andb in the·value space·,eithera < b orb < aora = b

·ordered· provides for:

• indicating whether an·order-relation· isdefined on a·value space·, and if so,whether that·order-relation· isa·partial order· or a·total order·

4.2.3 bounded

[Definition:]  A valueu in an·ordered· ·value space· Uis said to be aninclusive upper bound of a·value space· V(whereV is a subset ofU)if for allv inV,u >=v.

[Definition:]  A valueu in an·ordered· ·value space· Uis said to be anexclusive upper bound of a·value space· V(whereV is a subset ofU)if for allv inV,u >v.

[Definition:]  A valuel in an·ordered· ·value space· Lis said to be aninclusive lower bound of a·value space· V(whereV is a subset ofL)if for allv inV,l <=v.

[Definition:]  A valuel in an·ordered· ·value space· Lis said to be anexclusive lower bound of a·value space· V(whereV is a subset ofL)if for allv inV,l <v.

[Definition:]  A datatype isboundedif its·value space· has either an·inclusive upper bound· or an·exclusive upper bound·and either an·inclusive lower bound· or an·exclusive lower bound·.

·bounded· provides for:

• indicating whether a·value space· is·bounded·

4.2.4 cardinality

[Definition:]  Every·value space· has associated with it the concept ofcardinality. Some·value space·sare finite, some are countably infinite while still others couldconceivably be uncountably infinite (although no·value space·defined by this specification is uncountable infinite). A datatype issaid to have the cardinality of its·value space·.

Itis sometimes useful to categorize·value space·s(and hence, datatypes) as to their cardinality. There are twosignificant cases:

• ·value space·s that are finite
• ·value space·s that are countably infinite

·cardinality· provides for:

• indicating whether the·cardinality·of a·value space· isfinite orcountably infinite

4.2.5 numeric

[Definition:]  A datatype is said to benumeric if its values are conceptually quantities (in somemathematical number system).

[Definition:]  A datatype whose valuesare not·numeric· is said to benon-numeric.

·numeric· provides for:

• indicating whether a·value space· is·numeric·

4.3.1 length

[Definition:]  length is the numberofunits of length, whereunits of lengthvaries depending on the type that is being·derived· from.The value oflength ·must· be anonNegativeInteger.

Forstring and datatypes·derived· fromstring,length is measured in units ofcharacters as defined in[XML 1.0 (Second Edition)].ForanyURI,length is measured in units ofcharacters (as forstring).ForhexBinary andbase64Binary and datatypes·derived· from them,length is measured in octets (8 bits) of binary data.For datatypes·derived· by·list·,length is measured in number of list items.

·length· provides for:

• Constraining a·value space·to values with a specific number ofunits of length,whereunits of lengthvaries depending on{base type definition}.

<simpleType name='productCode'>   <restriction base='string'>     <length value='8' fixed='true'/>   </restriction></simpleType>

4.3.2 minLength

[Definition:]  minLength isthe minimum number ofunits of length, whereunits of length varies depending on the type that is being·derived· from.The value ofminLength ·must· be anonNegativeInteger.

Forstring and datatypes·derived· fromstring,minLength is measured in units ofcharacters as defined in[XML 1.0 (Second Edition)].ForhexBinary andbase64Binary and datatypes·derived· from them,minLength is measured in octets (8 bits) of binary data.For datatypes·derived· by·list·,minLength is measured in number of list items.

·minLength· provides for:

• Constraining a·value space·to values with at least a specific number ofunits of length,whereunits of lengthvaries depending on{base type definition}.

<simpleType name='non-empty-string'>  <restriction base='string'>    <minLength value='1'/>  </restriction></simpleType>

4.3.3 maxLength

[Definition:]  maxLength isthe maximum number ofunits of length, whereunits of length variesdepending on the type that is being·derived· from.The value ofmaxLength ·must· be anonNegativeInteger.

Forstring and datatypes·derived· fromstring,maxLength is measured in units ofcharacters as defined in[XML 1.0 (Second Edition)].ForhexBinary andbase64Binary and datatypes·derived· from them,maxLength is measured in octets (8 bits) of binary data.For datatypes·derived· by·list·,maxLength is measured in number of list items.

·maxLength· provides for:

• Constraining a·value space·to values with at most a specific number ofunits of length,whereunits of lengthvaries depending on{base type definition}.

<simpleType name='form-input'>  <restriction base='string'>    <maxLength value='50'/>  </restriction></simpleType>

4.3.4 pattern

[Definition:]  pattern is a constraint on the·value space· of a datatype which is achieved byconstraining the·lexical space· to literalswhich match a specific pattern. The value ofpattern ·must· be a·regular expression·.

·pattern· provides for:

• Constraining a·value space·to values that are denoted by literals which match a specific·regular expression·.

<simpleType name='better-us-zipcode'>  <restriction base='string'>    <pattern value='[0-9]{5}(-[0-9]{4})?'/>  </restriction></simpleType>