Well Formed

An XML document must follow simple syntax rules in order to be considered well formed. This stuff is so simple that it drives me batty that all these XML books and articles have to explain it all the time.

• XML Documents
• Elements v Attributes
• Basic Rules

XML Documents

An XML document consists of a prolog and a root element.

Prolog

A prolog is the section before the root element. It can consist of processing instructions, meta data items, or comments.

• Processing Instructions
  • It is common to see an XML declaration (<?xml version="1.0"?>) at the start of the first line of an XML document. Attributes such as encoding="UTF-8" standalone="yes" may be included in the XML declaration, but they must be put in that order: version, encoding, and then stand alone.
  • XML documents may reference a CSS (Cascading Stylesheet). EG: <?xml-stylesheet href="/style.css" type="text/css" title="default stylesheet"?>.
  • XML documents may reference an XSL (XML Stylesheet). EG: <?xml-stylesheet href="/style.xsl" type="text/xsl" title="default stylesheet"?>.
• Meta data items
  • XML documents may reference a DTD. EG: <!DOCTYPE note SYSTEM "InternalNote.dtd">.
• Comments
  • Just like HTML comments. EG: <!-- Comments -->.
  • Comments cannot occur within a tag. EG: <element <!-- This is illegal -->>.
  • Comments can occur in the prolog (the section before the root element).

Root Element

• An XML Document can have only one root element.
• A root element can have elements, comments, or CDATA comments.
• An element has some or all of these basic parts:
  • 1-2 tags. EGs: <one /> or <two></two>.
  • Element content. EG: <tag>Element content</tag>.
  • Attributes. EG: <tag AnAttribute="attribute value" />.
• Attributes of an element must not repeat. EG: <elt a="1" a="2"/> is not allowed.
• Here are some rules regarding element names.
  • Element names must begin with one of the following:
    • a letter
    • a colon (:)
    • an underscore (_)
  • Element names cannot begin with xml, in any combination of upper or lowercase.
  • After the initial character, the other characters cannot be any of the following:
    • a space
    • a tab
    • a non-printing character (EG: A carriage return or line feed)
    • an angle bracket

• Element relationships are frequently discussed using the usual family or tree metaphors. EG: It is easy to use the following xml section to identify parent, child, grandchild, grandparent, sibling, root, leaf, ancestors, descendants, etc.

  <a>
   <b>
     <c>
      <d/><e/><f/><f/>
     </c>
   </b>
  </a>

• Child elements of parent element can repeat. EG: The root element may have multiple child elements with the same name.
• Here is an example of a CDATA comment for content that contain characters that interfere with XML parsing (EG: < and &):

  <![CDATA[
  Information here is data for the XML document, but is ignored when it is parsed. 
  <anytag>EG this element is ignored</anytag>
  The tag following the next line is again part of the XML code.
  ]]>

Elements v Attributes

It is relatively simple to decide whether to put data in elements or in attributes of elements.

Note that some processes (like SQL Server and ADO)  default to putting database columns as attributes.

Basic Rules

There are a few basic rules that make a well-formed XML document.

• Elements can be nested but not overlapped. EG:

  <e>the quick<f>brown fox</f> jumped.</e> CORRECT!

  <e>the quick<f>brown fox</e></f> INCORRECT!

• Elements and attribute names are case sensitive. EG:

  <e>the quick.</e>
<e>the quick.</e> CORRECT!

  <E>the quick.</E>
<E>the quick.</E> CORRECT!

  <E>the quick</E>
<e>the quick.</e> INCORRECT!

  <e>the quick</E> INCORRECT!

• Non-empty elements require end tags. EG:

  <e>the quick.</e> CORRECT!

  <e>the <e>quick INCORRECT!

• Empty elements must have a beginning and an end tag or just a single tag ending with /> ( or  />  because some parsers/browsers prefer that).  EG:

  <e><e/>
<e />
<e/> CORRECT!

  <e>
<e> INCORRECT!

• Attributes values must be in quotes, either single or double. EG:

  <table rows="2"> CORRECT!

  <table rows=2> INCORRECT!

• Attributes cannot be minimized or empty. EG:

  <dl compact="compact"> CORRECT!

  <dl compact> INCORRECT!

• The allowed characters in XML are: tabs, line feeds, carriage returns, and legal Unicode characters excluding the surrogate blocks, FFFE, and FFFF.

  Char ::= #x9 | #xA | #xD | [#x20-#xD7FF] | [#xE000-#xFFFD] | [#x10000-#x10FFFF]

• XML has a smaller set of Character Entity References (CERs) that are the Predefined Character Entity References (PCERs) of XML. In contrast HTML has 252 CERs, while XHTML has 253 CERs.
  • &lt; (<)
  • &gt; (>)
  • &amp; (&)
  • &quot; (")
  • &apos; ('). Note that &apos; is defined in XML and XHTML but not in HTML.
  • You can add additional entity references but it is not a common practice. EG:

    <?xml version="1.0" encoding="UTF-8"?>
    <!DOCTYPE xsl:stylesheet [<!ENTITY nbsp "&#160;">]>

Page Modified: (Hand noted: 2007-10-04 19:25:55Z) (Auto noted: 2008-03-12 16:11:27Z)