CVS Revision $Revision: 1.6 $

Character Entity Files

The purpose of these files is to define character entities for later rendering when the text is viewed.  Publishers should provide a character entity file (often called something like charents.frag) with each text collection.  This file must define all character entities appearing in the collection.

The file's contents will look something like this:

<!ENTITY aacgr SDATA "&aacgr;" >
<!ENTITY aacute SDATA "&aacute; " >
<!ENTITY Aacute SDATA "&Aacute;" >
<!ENTITY Abar SDATA "&Abar;" >
<!ENTITY abar SDATA "&abar;" >
<!ENTITY Acirc SDATA "&Acirc;" >
<!ENTITY acirc SDATA "&acirc;" >
<!ENTITY Adot SDATA "&Adot;" >
<!ENTITY adot SDATA "&adot;" >
<!ENTITY AElig SDATA "&AElig;" >

After <!ENTITY, the first expression is the character entity, without an initial ampersand and a final semicolon.  This refers to character entities in the source files (so, for example, aacgr would appear in the source file as &aacgr;).  The second expression, in quotation marks, is the "definition" of the character entity for rendering it when it is displayed.

If SDATA does not appear, insert it in each line as above.

If unfamiliar character entities are used, it can be helpful to look to character entity files of other large collections.  If you can't find a definition, you may simply define it under the name of the character entity itself.  Just be aware that when the collection is served online, that entity will have to be rendered as a specific character.