XML - Grundlagen, Prinzipien und Anwendungen

Musterlösung Aufgabe 2 (XML DTD)

Verfeinern der DTD

Part 1 and 2 (Parameter Entity Delcarations and References and DTD Modularization)

Details on Modularizatin can be found in Chapter 4 of Learning XML.

Some important notes:

• An element type MUST NOT be declared more than once. XML standard. If you want to override an element declaration you need to use so-called conditional sections, cf. XML standard.
  
• Attributes can be declared more than once, i.e. overriding is possible; cf. XML standard. Note that the first declaration will take precedence!
  
• Entities can be declared more than once, i.e. overriding is possible, cf. XML standard. Note that the first declaration will take precedence!

DTD files that modularize (apart from a few modifications) the single DTD we developed for Exercise 1:

• tracklist1.dtd
  This file shows how to use Parameter Entity (PE) declarations in three ways: Firstly, we declare a list of common attributes length and year. These are referenced in the ATTLIST declarations of the elements track and album. Secondly, we define a new data type URI which is used to define the type of the attribute href for the element link. Lastly, we use the PE mechanism for declaring an enumeration of values through the entity declarations for styles and genders.
  
• address.dtd
  
• tracklist2.dtd
  Here, again we use the PE declarations and reference them. The referencing step acts as a simple text macro.

Part 3 (Character References)

Details on character encodings can be found in Chapter 9 of Learning XML, but to a short explanation is:

The character set for a document is the scheme by which characters are derived from the numerical values in its underlying form—e.g., a byte value of 252 corresponds to "ü". Now, in 1991 the Unicode consortium defined that 16 bits shall be used to describe all the international characters; i.e. 2^16 different characters are represented. To address a certain character, you need to use &#X; where X is the decimal or a hexadecimal number of the character, e.g., Ř maps to "Ř" Note, that the table, or mapping, of the unicode character set is *fixed*, e.g., the 252th character *always* corresponds to "ü". For compatibility reasons, the first 128 characters of the unicode set correspond to the US-ASCII character set, then the next 128 correspond to ISO-8859-1 (the latin set) and then the next 256 to...

The purpose of an "XML encoding" property is to allow the typing of the characters belonging the specified encoding scheme *directly* instead of referring to them as &#X;—i.e. to type an "ü" directly. Yet another point is that these decimal or hexadecimal values are bit hard to remember. Thus, it's convient to define another mapping: I.e. to use ü which maps to ü. But this mapping has to be defined first! In order that an XML parser knows this mapping, it has to be specified as a parameter entitiy, e.g.:

[
<!ENTITY % HTMLlat1 SYSTEM
     "http://www.w3.org/TR/xhtml1/DTD/xhtml-lat1.ent">
 %HTMLlat1;
]> 

Note that the file "xhtml-lat1.ent" contains definitions such as for instance

<!ENTITY uuml "&#252;">

Thus, if you use the parameter entity definition as shown above, you can refer to the entity definitions as a "normal" entity with "&". Therefore, to have an "ü" displayed, you can either use ü or if you used the parameter entity from above as ü

PS: If you "browse" such an XML file with Internet Explorer 6, it shows the "ü" correctly in both cases (i.e. with ü and ü). However, doing so with Netscape, Mozilla, or Opera causes an error :-(

XML Files with different character encodings:

• tracklist-ascii.xml
  This file uses US-ASCII encoding. Hence, it can use the german umlauts (or any other characters outside the ASCII character set) only through character references. In this case, we include an XHTML entity set that defines the special characters and reference the umlaut characters using the entity reference mechanism.
  
• tracklist-iso-8859-1.xml
  This file use the ISO-8859-1 encoding. The german umlaut characters are part of these character sets and thus can be included literally in this document.
  
• tracklist-utf-8.xml
  This file uses the "standard" XML UTF-8 character encoding. The german umlaut characters are part of these character sets and thus can be included literally in this document.

please send comments to xml-vl@dret.net last modification on Friday, 21-Apr-2006 14:28:17 CEST