The nsIDocCharset interface allows the user to view, 
 hint, and set which converter is used to read the document's 
 data.

 <P>Inside Mozilla all text is encoded as Unicode. By standardizing  
 on Unicode every piece of code that handles characters no longer 
 needs to: 

     Have access to a character encoding parameter.
     Ask is the data stateful?
     Ask is the data single byte?
     Ask is the data multi-byte?
     If multi-byte: have access to a routine that can
         determine how many bytes in the current character.

 When the document is read in a converter is used to convert 
 the document's data to Unicode.

 nsIDocCharset provides the ability to:

   See what converter was used when inputting the documents
   data.

   Override the character set for documents where the specified 
   fallback, or auto-detected character set is incorrect
[scriptable, uuid(9c18bb4e-1dd1-11b2-bf91-9cc82c275823)]
interface nsIDocCharset : nsISupports

Attributes

 Get/sets the encoding (converter) used to read the 
 document. Get returns the encoding used. Set forces 
 (overrides) the encoding. After forcing the charset the 
 embedding application will need to cause the data to be 
 reparsed in order to update the DOM / display.

 A force also sets the fallback encoding for this frame.
attribute string charset