Glossary

abstract syntax.

A term for the grammatical description of a language in which precedence relationships between the productions are ignored for the sake of simplicity. Such a description has fewer constraints than those specified by a concrete syntax.

anonymous type.

An unnamed type in XML Schema. The scope of the type is limited to the element or attribute within which it is defined. This is convenient for "one-off," in-place type definitions that don't need to be reused.

atomic type.

An XML Schema simple type that is not a list type or a union type.

atomic value.

The value of an item in the value space of an atomic type.

atomization.

A type-conversion process used to convert certain types of expressions to a sequence of atomic values. The main use is to extract atomic values from a node sequence, as in the application of an arithmetic function such as avg() or max() to an XPath expression yielding nodes. A type error is raised if the attempted conversion is illegal.

attribute.

One of two basic mechanisms in XML for partitioning data, the other being the element. The question of which should be used when is under debate. The common wisdom is that elements are used to encapsulate basic information; attributes are used to encapsulate "metadata" about that information.

Basic XQuery.

A description of the " core level" features and behavior expected of a conforming XQuery implementation. The optional "Schema Import Feature" and "Static Typing Feature" modify the items in this Basic XQuery description. For example, the use of a SchemaImport statement in basic XQuery raises a static error. Specifying that the XQuery processor implements a "Schema Import Feature" removes that limitation (among others).

BLOB.

In SQL, a binary large object.

canonicalized XML.

XML that has been canonicalized; that is, transformed into a standardized form on application of an agreed-upon set of rules that preserve an equivalence to the original document.

case-sensitive.

The property of a language specifying whether the distinction between upper- and lowercase characters is significant. XQuery is a case-sensitive language. All keywords use lowercase characters .

CLOB.

In SQL, a character large object.

closure.

A desirable property of XQuery; under closure, all expressions in the language return values that conform to those specified in the Query data model.

collection.

A sequence of nodes identified by a URI.

collation.

A language identifier and its set of character-mapping descriptions, used to determine comparison and sorting order.

Common Language Run-time.

A portable language run-time for a variety of Microsoft language environments such as Visual Basic, C#, J#, and C++.

complex type.

In XML Schema, an element type that may contain subelements and/or attributes (as opposed to a simple type).

composition technique.

As described in Chapter 6, an explicit or implicit mechanism used to determine the appearance of the XML for a set of relational tables.

compositionality.

A basic principle of functional languages such as XQuery, in which the output of any expression can be plugged into the input of any other with complete generality (with some possible semantic restrictions).

conclusion.

The "below the line" judgment in a formal semantics inference rule. When all the "above the line" judgments , or hypotheses, are true, then the conclusion is true by definition.

concrete syntax.

The grammatical description of a language that includes precedence and parentheses. The opposite of abstract syntax.

consolidation.

As described in Chapter 6, the process of reusing the same relational table during shredding to store data that occurs in multiple places.

constructor.

An XQuery mechanism for creating and returning new XML structures from within a query.

context item.

The item currently being processed during evaluation of the expression E1/E2 or E1[E2]. During the processing of E1/E2 , for example, E1 is first evaluated to a node sequence. Each node in turn is then set as the context item, and E2 is evaluated in respect to that context item.

data model.

For XML in general, this refers to the data model described by the XML Infoset [INFOSET], an abstract specification of the legal contents of an XML document. In this book this more often refers to the XQuery data model, an abstraction of the Infoset ().

See also [Query data model]

data-oriented (or data-centric) XML.

The set of XML vocabularies that represents data that is more tightly structured than in document-oriented (or document-centric) XML, such as an invoice, purchase order, or some other type of business data.

default mapping.

As described in Chapter 6, one of three composition techniques. The database is composed to XML based on a fixed mapping between column names and a final XML structure. The user has little or no control over the generated XML.

dependent join.

A join algorithm, as described in Chapter 6, in which tuples from one source are scanned, and for each tuple, another source is scanned to retrieve matching tuples.

derivation by extension.

In XML Schema, the mechanism of defining new types in terms of an existing base type, in which additional elements or attribute content are added to the base type.

derivation by restriction.

In XML Schema, the mechanism of defining new types in terms of an existing base type, in which the original value range is narrowed or further restricted.

derived type.

In XML Schema, the definition of a new type in terms of an existing datatype.

document order.

The order in which nodes are encountered as an XML file is read, either by a human visually or by an XML parser programatically.

Document Type Definition ( DTD ) .

The syntactic unit in the original XML Recommendation for specifying constraints on the names and numbers of elements and attributes allowed in a document. A fairly primitive but simple mechanism, it has been supplanted, where the need for greater specificity exists, by more complex mechanisms such as XML Schema and RELAX-NG.

document-oriented (or document-centric) XML.

The set of XML vocabularies that are used when authoring loosely structured natural language documents. DocBook is an example of one such vocabulary.

Document Object Model ( DOM ) .

A W3C-specified API for traversing XML documents.

dynamic error.

Dynamic errors occur during evaluation of a query when an error other than a static error is encountered. This causes implicit invocation of the fn:error() function, which can also be invoked explicitly in a query if desired.

dynamic semantics.

Specifies precisely the relationship between input data, an XQuery expression, and output data. Generally contrasted with static semantics.

See also [type soundness theorem]

dynamic type.

The type of an operand that is determined at run-time. This is the most specific type that can be determined for the value under consideration. The static type of the operand, on the other hand, is a compile-time property of an expression.

edge table.

A construct that represents the structure of a tree or graph in tabular form, in which each row identifies the properties of an edge, such as its name and the source and target vertices. The construct can be used to represent parent-child connections in an XML tree in a tabular, relational format. When representing trees, the edge table is isomorphic to the node table.

See also [node table]

Effective Boolean Value ( EBV ) .

The result of the process during evaluation of a sequence of items that first reduces the sequence to a single Boolean value. This is required in the evaluation of certain XQuery constructs, such as logical operations and conditionals. The XQuery 1.0 document specifies how EBVs are to be calculated, depending on the expression type. For example, before the logical expression sequenceA and sequenceB can be evaluated, each operand needs to be replaced by its equivalent EBV. If either sequence is empty, its EBV is false . If either sequence is a single Boolean false , its EBV is false , and so forth.

element.

One of seven node types in the data model, and generally the most important for containing information. Elements can optionally contain attributes, other elements, and text.

element constructor.

An XQuery syntactic structure that allows the creation of new XML element nodes directly in a query. The created elements may be specified either as literal "angle bracket " XML, or they may be computed dynamically.

element declaration.

The formal declaration of an element's allowable contents (or alternatively, its type), either through a DTD or XML Schema declaration.

encoding.

The physical mapping between the representation of characters in an XML document and their underlying Unicode basis.

end tag.

The angle-bracket representation in serialized XML that marks the end boundary of an element. In the element, for example, <elemName>contents</elemName> , </elemName> is the end tag.

environment.

Very generally, a mapping from symbols to values, the symbols and values depending on the context in which the environment is used. During static analysis, for example ( ), an "element name" symbol is mapped to its type definition, and a variable's name is mapped to its type. Dynamic evaluation uses the static environments, as well as mappings from a variable's name to its value, and so forth. Sometimes referred to as a "dictionary" or "context."

See also [formal semantics]

evaluation judgment.

A statement in the inference language of the formal semantics which states that the evaluation of an expression yields a specified value.

existential quantification.

An operation in which certain XQuery evaluations automatically (or implicitly) iterate over a sequence of items in order to evaluate an expression. For example, during evaluation of the path expression, //book[author = "Mark Twain"] , a number of author nodes may be present for each book node. Existential quantification occurs during evaluation of each such author sequence, in which the = operation will be determined as true if any of the author nodes have the content "Mark Twain" .

Extensible Markup Language ( XML ) .

Refers to both the precise syntactic definition of a markup language in which "tags" delimited by angle brackets impose a treelike structure on a document (see [XML]), and an alternative specification (see [INFOSET]) that describes the same structures much more abstractly.

facet.

A type of attribute in XML Schema nomenclature that acts to characterize or constrain the properties of a datatype. For example, minOccurs and maxOccurs facets can be used to constrain repetitions, and a combination of the two can represent the full usage of the occurrence indicators, * , + , and ? in DTDs.

factored type.

A union of item types and an occurrence indicator. For example, the factored type (element (plumber) element (surgeon))* denotes zero-or-more plumber or surgeon elements. Factored types make it easier for the XQuery type system to assign types to any expression that iterates over a sequence or that changes the order of a sequence (e.g., the for , some , every expressions and the function fn:distinct-nodes() ). Any XQuery type can be re- expressed as a factored type.

FLWOR expression.

A powerful and useful XQuery construct that allows both iterative and groupwise processing of sequences. It stands for for , let , where , order by , and return , which are the various clauses it contains, some optionally.

foreign key.

In relational database parlance, a column that represents, after normalization, data whose primary key is found in another table. For example, a university Class table might contain a foreign key column containing student IDs. These IDs in turn correspond to the primary student ID key in a separate Students table.

formal semantics.

The formal mathematical foundations that describe the operations and behavior of XQuery precisely, as described and defined in [XQ-FS].

function overloading.

The ability of a language processor (including XQuery) to support multiple functions having the same name but different signatures. The processor determines through function resolution which function to call, depending on the number and types of parameters passed.

function resolution.

The process of choosing the best function for a given function call, based on a static analysis of its argument types.

function signature.

The combination of function parameters and their types that is unique to each function, even those having the same name.

See also [function overloading]

functional language.

Somewhat of a misnomer, this has nothing to do with the definition of function in standard usage. It refers rather to a language that is primarily expression-based . Expressions can be freely composed together, the output resulting from evaluation of one expression becoming input to another. "Pure" functional languages typically do not have side effects, and every expression/operator can therefore be expressed as a function. XQuery is an "impure" functional language because it does have the side effect of creating new nodes with distinct identity.

general comparison operators.

The six XQuery operators = , != , > , >= , < , and <=. These act on sequences and perform existential quantification on their operands. For example, the comparison sequenceA > sequenceB returns true if any single item in sequenceA is greater than any single item in sequenceB .

global element declarations.

A form of declaration of an element type in both XML Schema and the XQuery type system in which the element is declared globally, not by reference to a local type declaration, so that the element is globally accessible within the schema and within any schema that imports the schema containing the declaration.

grouping.

The process of forming various data into groups, often applying aggregation functions such as count() and avg() to the assembly. FLWOR expressions are commonly used for this purpose, along with element constructors to wrap the resulting grouped data.

hierarchical join.

A query or mapping construct that creates hierarchical data from relational data by nesting matching records within a containing record.

hypotheses.

The "above the line" judgments in a formal semantics inference rule. When these are all true, then the conclusion (the "below the line" judgment) is true by definition.

identity.

Two nodes have the same identity only if they are exactly the same node.

indexing.

The process of adding secondary, "behind the scenes" information on where data is located, so that it can be efficiently found and retrieved at query time.

information item.

The Query data model builds on the definitions of the XML Information Set [INFOSET]. Information items are basic constituent units of the latter. For example, each element in a document is defined by an element information item, which in turn is defined by optional child information items and attribute information items, and so on.

Information Set ( Infoset ) .

An abstract definition of the constituent pieces of a well- formed XML document (see [INFOSET]), as opposed to the syntactic, angle-bracket definition in the original W3C Recommendation, Extensible Markup Language (XML) [XML].

inheritance.

In XML Schema, two mechanisms for defining new types in terms of existing types, called derivation by extension and derivation by restriction.

inlining.

As described in Chapter 6, a process in which children are placed into the same table as their parent during the process of decomposing (shredding) XML into relational tables.

inner join.

The selection of nodes from different documents (more correctly, different sequences of items) in which items are returned only if both sequences contribute a data point.

item.

The formal term for the individual members of a sequence in the Query data model, each of which is either a node or an atomic value.

item type.

The type of an item in the XQuery data model: either an atomic type, an element type, a comment node type, or a processing instruction type.

Java Virtual Machine.

A portable language runtime for the Java language.

join.

A concept borrowed from SQL, in which a FLWOR expression is used to combine information, most typically from two different documents. More formally , a join consists of selecting specific portions of the Cartesian cross-product of two or more sequences.

keyword.

Fixed terms specific to the XQuery language, such as if , and , for , where , and the like. These terms are not reserved, in the sense that elements and attributes, for example, can also use the same names. A query processor is able to distinguish between these usages by context through the structuring of the formal grammar.

lax.

One of three validation modes for checking an element against its XML Schema declaration. If a schema declaration is present, lax validation is the same as strict . If no declaration is present, lax acts as skip validation does.

list type.

One of the three XML Schema datatypes collectively called simple types. A list type is a sequence of atomic types.

LOB.

In SQL, an acronym for large object. LOBs are more specifically either character LOBs (CLOBs) or binary LOBs (BLOBs).

local element declaration.

A form of declaration of an element type in both XML Schema and the XQuery type system in which the element is declared locally, and not by reference to a global type declaration.

markup data.

A reference in Chapter 7 to data that is characterized by free- flowing text, with interspersed markup used to call out document structure and highlight important aspects of the text. Also known as loosely structured data; most document-centric usage scenarios refer to instances of this type of data.

matching judgment.

A statement in the inference language of the formal semantics that specifies the type of a particular value.

metadata.

Information about information, the form of which depends on the context. In a relational database, for example, metadata represents information on column names and types in the form of a catalog or schema. This is opposed to the tabular content itself. In XML, metadata is the structural information imparted by the markup itself, again as opposed to its content. In a slightly different usage, it refers to additional descriptive information about content that is often placed in attributes. In the following example,

 <lengthOfBody units="feet">42</lengthOfBody> 

the attribute units can be viewed as metadata that further describes the contents of the element.

mutator.

A stored procedure or method that changes a data value via a side effect.

namespace.

A mechanism described by the W3C Namespace Recommendation [NAMESP] that allows XML vocabulary sets to be reused. They can be imported into and coexist within single documents without collisions between overlapping element and attribute names. Names belonging to a particular namespace are prefixed by a namespace prefix and colon . The namespace prefix in turn is a shorthand for a fuller URL that represents the namespace.

namespace prefix.

An optional NCName prefix to an element or attribute name, prepended to it by a colon. The prefix is mapped to a unique namespace URL by a namespace declaration in the document.

nesting.

XML imposes a treelike structure on documents, in which elements may in turn contain other elements. These latter are also called subelements, or children, of their containing parent and are said to be nested within them.

node.

Collectively describes the seven types of data that can comprise an XML tree. Node types in the XML Infoset (and by extension the Query data model) are elements, attributes, text, comments, processing instructions, namespace nodes, and a document node representing the document itself.

node hierarchy.

The description of positional relationships between nodes in an XML tree. Most commonly, this refers to the superordinate-subordinate relationships between parent nodes and their children.

node table.

A construct that represents the structure of a tree in tabular form, in which each row identifies the properties of a node, such as its name, its value, and a reference to its parent node. The construct can be used to represent parent-child connections in an XML tree in a tabular, relational format. Since a node table represents a tree, it is isomorphic to the edge table.

See also [edge table]

node type.

One of the seven types of XML nodes in the XQuery data model: document node type, element node type, attribute node type, namespace node type, processing-instruction node type, comment node type, and text node type.

nonlinear hierarchy.

As described in Chapter 6, the result of a composition technique in which an element has multiple children that have been added to it through different hierarchical joins.

normalization rule.

The description in the formal semantics of the rules that specify how to transform the full XQuery language into a smaller core language that is easier to define, implement, and optimize.

null.

A special "out of band " value in SQL representing unknown or inapplicable data.

occurrence indicator.

A grammatical shorthand for specifying constraints on the allowable numbers of an item (used in a nontechnical sense). The indicator specifies whether the items are optional and/or how many times they're allowed to be repeated. The indicators follow:

? : optional (zero or one only)

* : optional (zero or more)

+ : one or more

These can be used to modify SequenceType expressions (e.g., element(*, po:address)+ ). They can also be found in the description of function parameters, and several type- related expressions such as instance of and typeswitch .

online transaction processing ( OLTP ) .

A form of data processing in which end-user transactions update databases in real time.

operational semantics.

The name in the field of logic and programming language theory for the techniques employed in the formal semantics to define XQuery. These include the normalization rules that define a language in terms of a smaller core, inference rules, and judgments.

optimization.

A process in which an XQuery processor will programmatically simplify the form of a query to make data retrieval more efficient.

orderspec.

A specification that states how two items are to be compared for the purpose of determining sort order. An order by clause may have one or more orderspecs.

outer join.

The selection of nodes from different documents (more correctly, different sequences of items) in which items are returned regardless of whether both sequences contributed a data point.

partial updates.

Updating only a few data values within a larger set of values in a document.

path expression.

The navigational portion of the XPath and XQuery languages, which specifies how to locate various nodes within the XML being queried.

Post-Schema Validation Infoset ( PSVI ) .

An extended version of the XML Infoset employed during XML Schema processing both to add type-related information and to normalize certain values.

predicate.

A modifier on an XPath step that specifies a filter against which items may be tested . Predicates can be numeric or conditional. Numeric predicates are used as an array index into a sequence to select a particular item; with conditional predicates, a Boolean test is applied against each item in the sequence.

primitive type.

The term for one of the nineteen XML Schema "base" datatypes, which are not derived from another type. Primitive types include xs:string , xs:boolean , xs:decimal , xs:float , xs:double , xs:QName , xs:hexBinary , xs:anyURI , hexadecimal and base 64 encodings for binary data, and nine datatypes representing time- and date-related information.

prolog.

The optional frontmost section of a query, which provides information on its context through various types of declarations and definitions. User-defined functions are defined here; for example; they can be referenced in the body of the query. Namespaces are declared in the prolog, as is information on collations and imported schemas to be used.

proof tree.

The description of a series of inference rules applied to an expression, each of which is true, which demonstrates the corresponding conclusion.

qualified name ( QName ) .

Taken from the XML Recommendation [XML], a QName represents a valid name type for several different XQuery constructs. Most commonly used to denote legal element and attribute names with optional namespace prefixes, QNames are also used to specify the names of datatypes, variables , and functions.

Query (or XQuery) data model.

The form of data model specific to XQuery. It's an abstraction of the Infoset [INFOSET], which ignores some detail, such as individual character information items and entity references, while adding type information supplied by XML Schema and the notion of multiple documents.

query language.

A language used to interrogate the contents of a data store.

Quilt.

A precursor to XQuery that provided much of its present functionality and syntax.

rank.

A numeric value assigned to the individual results of a full-text search. Usually provides some measure of the weight of the result with respect to the original query.

recursive function.

A function that calls itself. The canonical example is the definition of factorial as a recursive function: f(n) = n*f(n “ 1 ), where n > 0 and f(1) = 1 . XQuery supports recursive functions.

relational database.

The most common form of database, in which information is placed in tables, where rows hold multiple values, one value per column. The most common language for specifying queries against such data is SQL.

relational data model.

The data model underlying relational databases, introduced by E. F. Codd in 1970. Data is described as a sequence of tuples, or rows.

reserved word.

A keyword that can only be used in very specific syntactic contexts. XQuery has keywords ( for , where , etc.), but they're not reserved.

root node.

The topmost node of an XML tree: the document node in a full XQuery document; otherwise the single, topmost node of a fragment.

schema.

In general, a formal description of the names, types, and allowed patterns of occurrence of elements and attributes in a document. DTDs are a simple form of schema, as are the catalogs in a relational database. In the context of XQuery, the term generally refers to a schema as defined in the XML Schema Recommendation [SCHEMA].

schema import.

An optional XQuery feature that allows an XML Schema to be imported so that its definitions are available to a query.

schema validation.

The process in which information contained in an XML Schema is compared to data nodes to determine their type. If the node conforms to a schema declaration, it is considered valid, and its data model information item is "annotated," or marked , with the type information. Otherwise validation fails, and an error is reported .

semantic constraints.

Syntactically, the operands of any XQuery expression can be any other expression. This doesn't always make sense semantically, however. For example, while the XQuery grammar allows two node sequences to be added together (as a consequence of the principle of compositionality), this will actually produce an error at runtime.

semi-structured.

XML data in which the structure provided by the markup is not totally regular and has some variation in substructure between instances. In practice, such substructure can be denoted by the use of XML Schema union types, xs:choice annotations, or repeating groups. In general, the term semi-structured refers to any data model that is self-describing and may have arbitrary structure.

sequence type.

The XQuery syntax used to describe the type of a sequence. For example, a function parameter might be described as type node()+ , meaning one or more nodes (of any of the allowable seven node types), or element(LastName, nameType)? , meaning an optional LastName element conforming to a schema definition for nameType .

shredding.

The name for the process in which XML data is decomposed into values suitable for storage in relational tables.

simple type.

An XML Schema primitive type, or a type derived from an XML Schema primitive type, by restriction, list, or union.

skip.

One of three validation modes for XML Schema. The element being validated must be well-formed. It and all subordinate elements are labeled as xs:anyType , and any attributes are labeled xs:anySimpleType .

SOAP.

An acronym for Simple Object Access Protocol, an XML-based protocol for web services.

sort keys.

The operands of an order by expression that optionally modifies a FLWOR expression. The key or keys specify the ordering mechanism by which the output of the FLWOR is sorted.

Sorted Outer Union.

A query or data-mapping construct that creates hierarchical data from relational data by grouping the relational data on common keys.

SQL.

Originally an acronym for Structured Query Language, SQL is a language for defining, modifying, and controlling the data in a relational database. Pronounced officially with the individual letters sounded out, as S-Q-L; frequently as "sequel" informally.

SQL-2003.

The most recent ISO Standard version of SQL.

SQL/XML.

An ISO proposal that adds standard XML extensions to SQL. It adds the definition of a native XML datatype to the language, which allows XML documents to be treated as relational values in columns . It also provides implicit mappings from relational data to XML.

start tag.

The angle-bracketed portion of XML markup that denotes the point in a serialized document where an element begins. Start tags can optionally contain attributes.

static environment.

All the information that is available during static analysis of a query, prior to its evaluation. This information comes from the query prolog and from the host environment.

static semantics.

The portion of the formal semantics that describes how static types are associated with expressions in XQuery.

static type.

The type of an expression as determined by the static typing rules applied during static analysis. Static analysis examines only the query and types used in the query (as specified by imported XML schemas) and is independent of any input data.

static type-checking.

The process of checking during static analysis that an inferred static type is a subtype of an expected type. This guarantees type soundness during dynamic execution since all type errors are deleted statically.

step.

The part of an XPath expression that generates a sequence of nodes, optionally filtered by a predicate. A step always returns a sequence of nodes in document order and eliminates duplicate nodes based on node identity.

store.

A mapping in the formal semantics relating node identifiers to node values.

strict.

One of the three validation modes under XML Schema. A schema declaration must be present, and the element must validate with respect to that declaration. The element and all subordinate content are labeled with the appropriate type annotation.

string value.

The content of a node displayed as a single string. The string-value() accessor ( ) determines how this is derived for each specific node type. For example, the string value of a document node is defined as the concatenation of the string values of all the text node descendants of the document in document order. The string value of an element is similarly the concatenation of the string values of all the text node descendants of the element, and the string value of an attribute is simply its value.

See also [data model]

structured data.

Data that fits easily into a predefined, homogenous model.

See also [semi-structured]

subelement.

An element contained within another, superordinate element.

substitution group .

A mechanism in XML Schema that allows elements of one name to be substituted for elements of another name.

tag.

The angle-bracket- enclosed section of XML markup that denotes either the start or the end point of an element. Can be one and the same for an empty element.

three-valued logic.

A form of logic in SQL that specifies the results of logical operations on operands having the standard binary values of true and false , as well as on operands representing the special null value. Not used in XQuery.

transitivity.

A mathematical property of an operator that states, given that operandA op operandB and operandB op operandC are both true, that operandA op operandC is also true. XQuery's value comparisons are transitive (because $var1 eq $var2 and $var2 eq $var3 implies $var1 eq $var3 ), while general comparisons are not (since $book1/author = $book2/author and $book2/author = $book3/author does not imply that $book1/author = $book3/author ). The latter fails transitivity because existential quantification is applied to the operands of general comparisons and may affect each pair differently.

trigger.

A procedure that is automatically executed in a DBMS in response to certain operations, such as update of a particular column.

two-valued logic.

Standard binary logic in which a Boolean operation performed on its operands returns either true or false .

type.

A set of constraints, as specified in an XML schema, that denotes a set of values in the XML data model.

type annotation.

The property of element and attribute nodes in the Query data model that describes the type of the node, as determined by the schema-validation process.

type error.

A type error is raised by XQuery when the type of an expression does not match the type required by the context in which it appears. For example, passing a string as a function argument where an integer is expected would raise a type error. How type errors are reported is implementation-dependent.

type promotion.

A process in which atomic values are promoted from one type to another, both during the resolution of function calls and the processing of operators that require numeric operands. For example, if a function parameter is declared to be of type xs:float , the function can actually be passed an argument of type xs:decimal , which will be promoted to type xs:float .

type soundness.

The property of a type system that guarantees that evaluation of a type-correct expression will never raise a type error.

type soundness theorem.

Expressed in two parts , this theorem provides the mapping between static and dynamic semantics in the formal semantics. The first part states that static analysis and dynamic evaluation in matching static and dynamic environments yield matching results. The second states that if an expression has a particular static type, dynamic evaluation of that expression will never yield a type error, and if the expression yields a value, then the value matches the specified static type.

type system.

The XQuery type system consists of the set of XML Schema primitive and node types, the set of operators for combining and comparing types, the rules for associating types with values in the XQuery data model, and the rules for associating types with XQuery expressions.

typed value.

The simple-valued content of an attribute or element node whose type annotation denotes a well-formed node, a node with simple content, or a node with mixed content.

typing judgment.

The formal description, in the nomenclature of the formal semantics, that states the type of a particular expression in a static environment.

Unicode.

A unified system for representing international character sets. Standardized by both the ISO and the Unicode Consortium.

union type.

A type that denotes a choice of XML Schema datatypes. The value space and lexical space of the new type are the union of the value space and the lexical space of the types being unioned.

universal quantification.

Derived from the field of predicate logic, a form of existential semantics that posits that every member of a sequence meets a specified condition. More concretely in XQuery, a quantified expression of the form, every ... satisfies .

Universal Resource Identifier ( URI ) .

A standard mechanism for addressing Web-based resources. Primarily used in XQuery to identify namespaces, as well as the targets of the doc() and collection() input functions.

untyped data.

Data whose type is unknown because it's not described by a schema. XQuery associates special type annotations with untyped values, and operators apply special conversion rules to untyped data, depending on the context in which it is used.

update.

The ability to change the value of nodes and write the updated value back to the underlying data store from which it was selected in the first place. Not a part of XQuery Version 1.

use case.

A hypothetical usage scenario, specifying a specific query, its input data, and its expected results. Discussion of proposed new features within the Query working group has centered around such scenarios, which are formalized in the Use Cases document [XQ-UC].

validation.

In XML generally, the process of determining whether a document instance complies with the definitions and constraints specified by a DTD or schema. In XQuery, which uses XML Schema, a major by-product of the validation process is determination of the type of each node being validated.

validation context.

Controls how the name of a given element is matched against declarations in the imported schemas. For example, a validate expression can specify a context for locally declared elements or attributes.

validation mode.

Specified in the prolog, this can be one of the following: lax , strict , or skip .

value.

The contents of an actual data instance, as opposed to its type.

value comparison operators.

XQuery operators that are used to compare two single, non-node values. The general comparison operators, which are used to compare sequences of items, add the concept of existential semantics to build on this base.

value space.

The set of allowable values for a given datatype. For example, the value space for the built-in type xs:boolean consists of { true, false } . See http://www.w3.org/TR/xmlschema-2/#value-space.

virtual XML view.

A view of non-XML data that is presented in an XML format without actually materializing the view (or caching it for later reuse).

WebDAV.

An acronym for Web-based Distributed Authoring and Versioning. It is an IETF Proposed Standard, a set of extensions to the HTTP protocol that allows users to edit and manage files collaboratively on remote web servers.

whitespace.

Defined in the XML Recommendation [XML] as one or more spaces (#x20), carriage returns, line feeds, or tabs. Whitespace is meaningful in an XML document if it is not being validated against a DTD or schema. If it is being validated, its significance depends on the schema definition.

World Wide Web Consortium ( W3C ) .

The organization in charge of developing and maintaining the common protocols that define and govern the World Wide Web. Member companies sit on working groups that work by consensus to develop W3C Recommendations. Some of the technical areas under W3C purview are HTTP, HTML, PNG, and the rapidly growing umbrella of XML-related technologies.

xdt:dayTimeDuration.

An XQuery-specific specialization of XML Schema's duration datatype. Its lexical space contains only day, hour , minute, and second components , and its value space is the set of fractional second values.

xdt:untypedAtomic.

An atomic value for which no type information is available. An atomic value inside an untyped node, such as <answer>47</answer> is of type xdt:untypedAtomic .

xdt:yearMonthDuration.

An XQuery-specific specialization of XML Schema's duration datatype. Its lexical space contains only year and month components, and its value space is the set of integer months.

XML datatype.

A relational datatype in SQL for representing XML data.

XML fragment.

An XML tree whose root node is not a document node. A partial tree.

XML Schema.

The W3C specification that provides a more complete mechanism for specifying validity constraints on XML documents than that provided by the original XML mechanism of Document Type Definitions (DTDs).

XML storage fidelity.

A term used in Chapter 7 to describe the degree to which the storage of XML documents maintains properties of the original format. There are three variations: (1) string-level fidelity, in which storage preserves the original XML document, code-point for code-point; (2) Infoset-level fidelity, in which storage preserves the document structure on the level of the XML Infoset model; and (3) relational fidelity, in which storage preserves relationally important information and discards XML-specific properties such as document order.

XPath.

The W3C specification that acts as an enabling technology for other W3C technologies, such as XSLT and XQuery, in which it's embedded. It provides a mechanism for searching for specific nodes within an XML document, in which both structural and content constraints can be specified.

XQuery data model.

See [Query data model]

xsi:nil.

A special attribute defined in XML Schema that permits the element to be accepted as valid with no content during the process of Schema validation, even though the element has a declared type requiring content.

XSLT.

A competing and complementary mechanism to XQuery for locating and transforming XML from one format to another. XSLT uses a stylesheet mechanism to drive a transformation from an XML input tree to an output tree that can be either XML, HTML, or text.