XML's Idea of Content and Names

XML documents can use a wide array of characters. They are defined by Unicode, which currently (Version 5.0) defines more than 100'000 characters (#100'000 added in 2005).

<?xml version="1.0" encoding="UTF-8"?>
<JAPANESE>
 <TITLE>専門家リスト </TITLE>
 <ITEM>アシム・アブドゥラー氏（コマースネット事務局長）</ITEM>
 <ITEM>アラン・Ａ・メッコラー氏（メッコラーメディア会長兼ＣＥＯ）</ITEM>
 <ITEM>アラン・サルディッチ氏（メトリコムディレクター）</ITEM>
 <ITEM>ウィスター・ウォルコット氏（パイロットネットワーク・サービシズ副社長）</ITEM>
 <ITEM>・エリック・リンゲワルド氏（ビー・インク副社長）</ITEM>
 <ITEM>ジェームス・Ｌ・バークスデール氏（ネットスケープ・コミュニケーションズ社長）</ITEM>
</JAPANESE>japanese1.xml

<?xml version="1.0" encoding="UTF-8"?>
<文書 改訂日付="１９９９年３月１日">
 <題>サンプル</題>
 <段落>これはサンプル文書です。</段落>
 <!-- コメント -->
 <段落>会社名</段落>
 <図面 図面実体名="サンプル" />
</文書>japanese2.xml

XML and Unicode

• XML is based on Unicode
  • XML is defined in terms of character structures
  • how these characters are encoded is not part of XML
• How are XML documents encoded?
  • applications can use any character encoding they like
  • XML processors must support UTF-8 and UTF-16
  • XML processors may support any number of additional encodings
• How is the encoding encoded?
  • part of the XML document: <?xml version="1.0" encoding="UTF-8"?>
  • bootstrap problem solved heuristically or by out-of-band information

Identifiers are Essential

• Uniform Resource Locator (URL) is the old concept
  • introduced to distinguish between locating and naming
  • locating and naming are two ways of identification
  • URLs have been replaced by URIs, technically URLs do not exist anymore
• URIs identify resources
  • some resources may be retrieved using a protocol: http://dret.net/netdret/
  • not all resource access is retrieval: mailto:dret@berkeley.edu
  • sometimes computers are not required: tel:+1-510-6432253
  • or resources cannot be located: urn:ietf:rfc:2648
  • or location is the only means of identification: geo:27.988056;86.925278

Markup?

• Structures are encoded using special characters
  • a fundamental difference when comparing to binary formats
  • markup languages can be read and modified using text-based tools
  • programs must treat markup characters in a special way
• Documents are content interspersed with markup (i.e., structures)
  • XML-aware software interprets the markup
  • XML-unaware software just sees a text file
  • modifications must be made XML-aware (e.g., inserting AT&T as AT&T)
• You have to pay the The Price for Markup

Basic Concepts

• XML Documents have an XML declaration (optional)
• There is exactly one document element (a.k.a. root element)
• Elements may be nested (there is no conceptual limit)
  • elements may be repeated (they can be identified by position)
• Elements are marked up using tags
  • most elements have content, surrounded by start and end tags
  • empty elements are allowed and may use a special notation
• Elements may have attributes (zero to any number)
  • attributes can only occur once on an element (i.e., they cannot be repeated)

<?xml version="1.0" encoding="UTF-8"?>
<element>
 <subelement attribute="value">Content</subelement>
 <subelement a2="value2">More Content</subelement>
 <empty-element a3="v3"></empty-element>
 <empty-element a4="v4" a5="v5"/>
</element>my-first.xml

Tree Syntax

• Markup is important, but only a notation
• XML documents are trees with different node types
  • nodes so far: document, element, attribute, text
  

Elements

• Elements can use a wide variety of names
  • Allowed: html, id9832798472, _, :, こんにちは
  • Disallowed: leading numbers, spaces, control characters
• Element names usually convey some information about the content
  • this is not reliable and highly language-dependent
  • it is very useful when working with a known vocabulary
  • it is potentially harmful when working with an unknown vocabulary
• Elements are the foundation for XML's versatility
  • they can be nested (<address><city>Berkeley</city><zip>94709</zip>…)
  • they can be repeated (<givenname>Erik</givenname><givenname>Thomas</givenname>)
  • their sequence can convey additional information (given names have a sequence)

Attributes

• Additional information pertaining to elements
• Traditionally, anything that is not considered content
  • SGML is a document markup language
  • XML uses SGML's concepts
  • XML has its roots in the document world
• Elements: Content (i.e., Data); Attributes: Metadata
• Documents often distinguish by what is textual content

 <section id="xml" author="bob">
  <title>Extensible Markup Language (XML)</title>
  <p>XML is based on SGML (Section <ref name="sgml"/>) ...</p>
  <p type="example">XML can be used ...</p>
  <section id="xml-syntax" author="dret">
   <title>XML Syntax</title>
   <p>Section <ref name="sgml-syntax"/> describes ...</p>
  </section>
 </section>section.xml (line 12-20)

Attribute Syntax

• Naming rules are the same as for Elements
• Attributes always appear within an element's start tag
• Attributes are name/value-pairs
  • the value is enclosed in single or double quotes
• Attribute with a single-quote value: elem attr="Single: '"/
• Attribute with a double-quote value: elem attr='Double :"'/
• How can attribute values contain both?

The Price for Markup

• Markup characters have a special meaning
  • < opens a tag
  • for attribute values, quotes delimit the value
• The literal use of a markup character requires escaping
  • XML's entities can refer to pieces of content
  • entity syntax is &name; for referring to the entity name
  • XML has 5 predefined entities: <, >, &, ', "
• Attribute using both kinds of quotes: <elem attr="Single ' and Double &quot;"/>

<li>Attribute using both kinds of quotes: <code>&lt;elem attr="Single ' and Double &amp;quot;"/></code></li>

Mixed Content

The term Mixed content in XML refers to elements which have text content mixed with elements. What these elements do depends on the elements , but the important point is that they are on the same level as the text nodes of the mixed content.

<p>The term <em>Mixed content</em> in XML refers to elements <a href="http://www.w3.org/TR/xml/#sec-mixed-content">which have text content mixed with elements</a>. What these elements do depends on the elements <img style="height : 1em" src="smiley.gif"/>, but the important point is that they are on the same level as the text nodes of the mixed content.</p>

Mixed Content Usage

• Database people find mixed content irritating
  • cannot be easily mapped to relational structures
  • is more document-like than data-like
  • much harder to optimize for query analysis and query processing
• Document people find mixed content very intriguing
  • textual content can still be used as simple text
  • markup provides additional information for rich text
  • start with a text-only document and use markup to add structure to it

Whitespace

• XML documents often are pretty-printed
• Whitespace text nodes often are not really content
  • XML whitespace characters are space, tab, newline, and carriage return
  • whitespace text nodes are text nodes containing only whitespace characters
  

Significant Whitespace

• Some whitespace text nodes are relevant
• Usually text nodes in mixed content elements

Whitespace can be very important!

<p>Whitespace <i>can be</i> <u>very</u> <b>important</b>!</p>

Observing XML Syntax

• XML's syntax requires you to use the right characters
  • the grammar alone allows many XML errors
  • additional constraints ensure that everything is used correctly
• XML processors (a.k.a. XML parsers) check for these rules
  • if there are problems, the document cannot be interpreted as XML
  • otherwise, the document is said to be well-formed
• Only well-formed documents can be regarded as a tree
  • other documents are not XML at all, even though they may be close
  • XML processors must report problems to the application (no silent recovery)

Validity

• Well-formed documents observe XML rules
  • they observe the XML syntax
  • they observe all well-formedness constraints
• Applications require the right elements and attributes
• Validity is a more comprehensive concept
• Valid documents observe additional rules
  • they must be well-formed documents
  • they must adhere to the constraints defined in a Document Type Definition (DTD)

Semantics

• XML is a language for encoding trees
  • Elements and attributes are labeled node in this tree
  • the labels can be chosen freely by document authors
• The tree's meaning is nothing XML is concerned with
  • peers must have a mutual understanding of the semantics
  • XML without mutual understanding is almost useless
  • reverse engineering often is possible, but it is risky and brittle