Text.XML.HXQ.XQuery

HXQ is a fast and space-efficient compiler from XQuery (the standard query language for XML) to embedded Haskell code. The translation is based on Haskell templates. It also provides an interpreter for evaluating ad-hoc XQueries read from input or from files and optional database connectivity using HDBC. For more information, look at http://lambda.uta.edu/HXQ/.

The XML Data Representation

type Prefix = String

Namespace prefix

type URI = String

Namespace URI

type LocalName = String

Local name

data QName

A qualified name has a namespace prefix, a URI, and a local name

Constructors

QName

Fields

prefix :: !Prefix

uri :: !URI

localName :: !LocalName

Instances

Eq QName

Ord QName

Show QName

type Attributes = [(QName, String)]

XML attributes are bindings from qualified names to values

data XTree

A rose tree representation of XML data. An XML element is: XElem tagname atributes preorder parent children. The preorder numbering is the document order of elements. The parent is a cyclic reference to the parent element.

Constructors

XElem !QName !Attributes !Int XTree [XTree]

an XML tree node (element)

XAttr !QName !String

attribute construction

XText !String

an XML tree leaf (PCDATA)

XInt !Int

an XML tree leaf (int)

XFloat !Double

an XML tree leaf (double)

XBool !Bool

an XML tree leaf (boolean)

XPI String String

processing instruction

XGERef String

general entity reference

XComment String

comment

XError String

error message

XNull

null value

XType Type

type information

XNoPad

marker for no padding in XSeq

Instances

Eq XTree

Show XTree

type XSeq = [XTree]

A sequence of XML fragments

type TVar = Int

A type variable

type TQualifier = Char

Type qualifier: *, +, or ?

data Type

An XQuery type

Constructors

TVariable !TVar

type variable (needed for polymorphic type inference)

TBase !QName

xs:integer, xs:string, ...

TItem !String

item(), node(), ...

TNamed !QName

reference to a user-defined type

TElement !String !Type

element tag { t }

TAttribute !String !Type

attribute name { t }

TAny

any element or attribute content

TEmpty

()

TSequence !Type !Type

t1, t2

TInterleaving !Type !Type

t1 & t2

TChoice !Type !Type

t1 | t2

TQualified !Type !TQualifier

t*, t+, or t?

Instances

Eq Type

Show Type

putXSeq :: XSeq -> IO ()

Print the XQuery result (which is a sequence of XML fragments) without buffering.

The XQuery Compiler

xq :: String -> Q Exp

Compile an XQuery that may perform IO (such as reading an XML document or calling a user function). When the compiled code is evaluated, it returns a value of type IO XSeq.

xe :: String -> Q Exp

Compile an XQuery expression that does not perform IO. When the compiled code is evaluated, it returns a value of type XSeq.

qx :: QuasiQuoter

Quasi-quotation for HXQ (for ghc 6.09 or later). For example, [qx| doc("data/cs.xml")//gpa |] is equivalent to xq "doc(\"data/cs.xml\")//gpa".

The XQuery Interpreter

xquery :: String -> IO XSeq

Evaluate the XQuery using the interpreter.

eval :: XSeq -> IO XSeq

The XQuery interpreter as an XQuery function.

The XQuery Command Line Interpreter

commandLineInterpreter

Arguments

:: [String]

program arguments

-> IO ()

The XQuery command line interpreter used by the main program (xquery). The program arguments may contain the following command line options:

xquery-file

Evaluate the XQuery code in xquery-file using the interpreter

-db database-name

Use the relational schema database-name during querying

-c xquery-file

Compile the XQuery code in xquery-file into Haskell code

-o haskell-file

Set the Haskell file for -c (default is Temp.hs)

-p XPath-query xml-file

Interpret the XPath query against the xml-file

-v

Print verbose information (the AST and the optimized plan)

-t

Print timing information

-tp

Print typing information (experimental)

Without an xquery-file, it reads and evaluates the input using the HXQ interpreter. The input may be a single XQuery or a declare variable or a declare function expression. To write an XQuery in multiple lines, wrap it in { }.

Validation using XML Schema

validateFile :: FilePath -> FilePath -> IO Bool

Validate the XML document against the XML Schema. Also done using the validate XQuery form.

The XQuery Compiler with Database Connectivity

xqdb :: String -> Q Exp

Compile an XQuery that may perform IO and/or queries a database. When the compiled code is evaluated, it returns Connection -> IO XSeq.

The XQuery Interpreter with Database Connectivity

xqueryDB :: String -> Connection -> IO XSeq

Evaluate the XQuery with database connectivity using the interpreter.

Shredding and Publishing XML Documents Using a Relational Database

type Path = [String]

XPath to reach a table/column

data Table

A relational schema representation

Constructors

Table String Path Bool [Table]

table-name relative-path mixed-content? components

Column String Path Int

column-name relative-path max-byte-size

Instances

Read Table

Show Table

genSchema

Arguments

:: Connection

database connection

-> FilePath

XML document pathname

-> String

schema name

-> [String]

excluded tags

-> IO Table

Create a schema for an XML document into the database under the given name. The excluded tags are HTML tags to be ignored

shred

Arguments

:: Connection

database connection

-> FilePath

XML document pathname

-> String

schema name

-> IO ()

Store an XML document into the database under the given name.

shredC

Arguments

:: String

database name

-> FilePath

XML document pathname

-> String

schema name

-> Q Exp

Store an XML document into the database under the given name. Generates Haskell code. It's 3 times faster than shred.

isSchema

Arguments

:: Connection

database connection

-> String

schema name

-> IO Bool

True if there is a relational schema stored in the database under the given name

printSchema

Arguments

:: Connection

database connection

-> String

schema name

-> IO ()

Print the relational schema stored in the database under the given name

createIndex

Arguments

:: Connection

database connection

-> String

schema name

-> String

the tag name of the elements to be indexed

-> IO ()

Create a secondary index on tagname for the shredded document under the given name..

Other Database Functions

connect

Arguments

:: String

database name

-> IO Connection

Connect to a relational database

disconnect

Arguments

:: Connection

database connection

-> IO ()

Disconnect from the relational database

commit

Arguments

:: Connection

database connection

-> IO ()

commit the updates to the database

rollback

Arguments

:: Connection

database connection

-> IO ()

rollback the updates from the database