XML and Databases

1
XML and Databases

• 198541
• Spring 2007

2
XML Motivation
3
XML Motivation

• Huge amounts of unstructured data on the web
  HTML documents
• No structure information
• Only format instructions (presentation)
• Integration of data from different sources
• Structural differences
• Closely related to semistructured data

4
Semistructured Data

• Integration of heterogeneous sources
• Data sources with non rigid structures
• Biological data
• Web data
• Need for more structural information than plain
  text, but less constraints on structure than in
  relational data

5
Characteristics of Semistructured Data

• Missing or additional tuples
• Multiple attributes
• Different types in different objects
• Heterogeneous collection
• Self-describing, irregular data with no apriori
  structure

6
HTML Document Example
Type of information

• lth1gt Bibliography lt/h1gt
• ltpgt ltigt Foundations of Databases lt/igt
• Abiteboul, Hull, Vianu
• ltbrgt Addison Wesley, 1995
• ltpgt ltigt Data on the Web lt/igt
• Abiteoul, Buneman, Suciu
• ltbrgt Morgan Kaufmann, 1999

Title
Authors
Year
book
7
The Idea Behind XML

• Easily support information exchange between
  applications / computers
• Reuse what worked in HTML
• Human readable
• Standard
• Easy to generate and read
• But allow arbitrary markup
• Uniform language for semistructured data
• Data Management

8
XML
9
XML

• eXtensible Markup Language
• Universal standard for documents and data
• Defined by W3C
• Set of emerging technologies
• XLink, XPointer, XSchema, DOM, SAX, XPath,
  XQuery,

10
XML

• XML gives a syntax, not a semantic
• XML defines the structure of a document, not how
  it is processed
• Separate structural information from format
  instructions

11
XML Example

• ltbibliographygt
• ltbookgt lttitlegt Foundations lt/titlegt
• ltauthorgt Abiteboul lt/authorgt
• ltauthorgt Hull lt/authorgt
• ltauthorgt Vianu lt/authorgt
• ltpublishergt Addison Wesley
  lt/publishergt
• ltyeargt 1995 lt/yeargt
• lt/bookgt
• lt/bibliographygt

12
XML Terminology

• Tags book, title, author,
• Start tag ltbookgt
• End Tag lt/bookgt
• Elements are nested
• Empty Element
• ltreviewsgtlt/reviewsgt gt ltreviews/gt
• XML Document single root element
• XML Document is well formed matching tags

13
XML Attributes

• Attributes are ltname, valuegt pairs that
  characterize an element.
• ltbook price 55 currency USDgt
• lttitlegt Foundations of Databases lt/titlegt
• ltauthorgt Abiteboul lt/authorgt
• ltyeargt 1995 lt/yeargt
• lt/bookgt
• Can define oid, but they are just syntax

14
More XML

• Text can be CDATA or PCDATA
• Entity References amp, gtgt,
• Processing Instructions lt?blink?gt
• Comments lt!-- comment text --gt

15
Well Formed XML Documents

• Elements must be properly nested
• ltbookgtlttitlegt Foundations of Databases
  lt/titlegtlt/bookgt
• But Not
• ltbookgtlttitlegt Foundations of Databases
  lt/bookgtlt/titlegt
• There must be a unique root element
• Elements can be of
• element content
• or mixed content
• lttitlegtThis is ltbgtMixedlt/bgtContentlt/titlegt

16
XML Potential

• Flexible enough to represent anything
• Stock market, DNA, Music, Chemicals
• Weather information
• Wireless network configuration
• Enables easy information exchange
• Between companies
• Within companies
• Standard everybody uses the same technology

17
XML Limitations

• XML is only a syntax for documents
• We need tools!
• Editors and parsers
• Programming APIs (for Java, C, etc.)
• Languages to manipulate XML (how many books?)
• Schemas (What is a book like?)
• Storage (What if you have a lot of XML?)
• Transfer protocols (How do you exchange it?)
• What about XML in Chinese?
• How can XML fit into my phone?
• Query processing?

18
XML Schema Language
19
DTDs Document Type Descriptors

• Similar to a schema
• Grammar describing constraints on document
  structure and content
• XML Documents can be validated against a DTD

lt!ELEMENT Book (title, author)gtlt!ELEMENT title
PCDATAgtlt!ELEMENT author (name, address,
age?)gtlt!ATTLIST Book id ID REQUIREDgtlt!ATTLIST
Book pub IDREF IMPLIEDgt
20
Shortcomings of DTDs

• Useful for documents, but not so good for data
• No support for structural re-use
• Object-oriented-like structures arent supported
• No support for data types
• Cant do data validation
• Can have a single key item (ID), but
• No support for multi-attribute keys
• No support for foreign keys (references to other
  keys)
• No constraints on IDREFs (reference only a
  Section)

21
XSchema

• In XML format
• Includes primitive data types (integers, strings,
  dates,)
• Supports value-based constraints (integers gt 100)
• Inheritance
• Foreign keys

22
Example of XSchema

• ltschema version1.0 xmlnshttp//www.w3.org/199
  9/XMLSchemagt
• ltelement nameauthor typestring /gt
• ltelement namedate type date /gt
• ltelement nameabstractgt
• lttypegt
• lt/typegt
• lt/elementgt
• ltelement namepapergt
• lttypegt
• ltattribute namekeywords typestring/gt
• ltelement refauthor minOccurs0
  maxOccurs /gt
• ltelement refdate /gt
• ltelement refabstract minOccurs0
  maxOccurs1 /gt
• ltelement refbody /gt
• lt/typegt
• lt/elementgt
• lt/schemagt

23
XML Storage
24
Storing XML Data

• Different approaches
• Storing as text
• Using RDBMS
• Using a native system
• Tailored for XML, (NATIX, Tamino, Ipedo, etc.)
• Performance of the various approaches
• depends on your application

25
Storing XML as Text

• Simple
• Easy to compress
• No updates
• Need to parse the document every time it is needed

26
Storing XML in RDBMS

• Uses existing RDBMS techniques
• Costly in space, takes time to reconstruct
  original document
• Example techniques
• Schema with 2 relations tag and value
• Schema with n relations 1 per element name

27
Accessing and Querying XML Data
28
XML as a Tree DOM

• DOM Document Object Model
• Class hierarchy serving as an API to XML trees
• Methods of those classes can be used to
  manipulate XML (e.g., Nodechild, Nodename)
• Can be used from Java, C to develop XML
  applications.
• Each node has an identity (i.e., a unique
  identifier) in the whole document

29
XML as a DOM Tree

• Class hierarchy(node, element attribute)

bibliography
book
book
title
author
publisher
year
author
author
Foundations of Databases
Abiteboul
Hull
Vianu
Addison Wesley
1995
30
XML as a Stream SAX

• XML document event stream. E.g.,
• Opening tag book
• Opening tag title
• Text Foundations of databases
• Closing tag title
• Opening tag author
• Etc.
• SAX allow you to associate actions with those
  events to build applications
• Very efficient since it corresponds to events
  during parsing, but not always sufficient.

31
XPath

• Language for navigating in an XML document (seen
  as a tree)
• One root node
• types of nodes root, element, text, attribute,
  comment,
• XPath expression defines navigation in the tree
  following axis child, descendant, parent,
  ancestor,

32
XPath Examples

• Find all the titles of all the books
• //book/title
• Find the title of all books written by Charles
  Dickens
• //bookauthorCharles Dickens/title
• Find the title of the first section in the
  second chapter in Great Expectations
• //booktitleGreat Expectations/chapter2/sect
  ion1/title
• Find the title of all sections that come after
  the second chapter in Great Expectations
• //booktitleGreat Expectations/chapter2/foll
  owingsection/title

33
Querying XML Data

• Need for a language to query XML data
• Should yield XML output
• Should support standard query operations
• No schema required
• Several work on an XML query language XML-QL,
  XQuery,..

34
XQuery

• XPath included in XQuery
• FLWR expressions for let where return

FOR x IN document("bib.xml")/bib/book WHERE
x/year gt 1995 RETURN x/title
Result lttitlegt abc lt/titlegt lttitlegt def
lt/titlegt lttitlegt ghi lt/titlegt
35
How to process XML Queries?

• Use indexes
• Need to identify nodes
• Need to know relations between nodes
• Labeling Schemes
• Dewey encoding
• Prefix-Postfix encoding
• Twigstack

36
Web Services
37
What are Web Services

• Programming interfaces for application to
  application communication on the Web
• platform-independent,
• language-independent
• object model-independent
• Possibility to activate methods on remote web
  servers (RPC)
• 2 main applications
• E-commerce
• Access to remote data

38
XML and Web Services

• Exchange of information between application is in
  XML
• Input and Result
• Use of SOAP to generate messages
• Descriptions of the web service functionality
  given in XML, according to the WSDL schema

Web Services standards use XML heavily
39
Conclusions

• XML a very active area
• Many research directions
• Many applications
• Standards not finalized yet
• XQuery
• XML Schema
• Web Services

40
Some Important XML Standards

• XSL/XSLT presentation and transformation
  standards
• RDF resource description framework (meta-info
  such as ratings, categorizations, etc.)
• XPath/XPointer/XLink standard for linking to
  documents and elements within
• Namespaces for resolving name clashes
• DOM Document Object Model for manipulating XML
  documents
• SAX Simple API for XML parsing

41
References

• XML
• http//www.w3.org/XML/
• Sudarshan S. Chawathe Describing and
  Manipulating XML Data. IEEE Data Engineering
  Bulletin 22(3)(1999)
• XML Standards
• http//www.w3.org/ (XSL, XPath, XSchema, DOM)
• Storing XML Data
• Daniela Florescu, Donald Kossmann Storing and
  Querying XML Data using an RDMBS. IEEE Data
  Engineering Bulletin 22(3)(1999)
• Hartmut Liefke, Dan Suciu XMILL An Efficient
  Compressor for XML Data. SIGMOD Conference 2000
• XQuery
• http//www.w3.org/TR/xquery/
• Peter Fankhauser XQuery Formal Semantics State
  and Challenges. SIGMOD Record 30(3)(2001)
• Web Services
• http//www.w3.org/2002/ws/