The symbols [6]Names and [8]Nmtokens do not appear to be used in any
other rule. Are they redundant?
The symbol [40]elementdecl consists of the following expression:
'<!ELEMENT' S %Name S (%S S)? %contentspec S? '>'
What is the purpose of the '(%S S)?' part of this rule?
The second line of the rule for [50]Mixed is:
| '(' S? %( '#PCDATA' ) S? ')'
I cannot understand the purpose of the inner brackets in this part of
the rule.
There is also little written about interpretation of line-ending
codes. Although the standard states that white space and line-ending
codes are ignored in element content, nothing is said regarding the
age old problem of line-ending codes in mixed content. For example, in
SGML we know that an initial and final record end is ignored, so:
<P>
Here is a paragraph.
</P>
is considered the same as:
<P>Here is a paragraph.</P>
Could we have some clarifications on this point?
Thanks.
Neil.
-----------------------------------------------
Neil Bradley - Author of The Concise SGML Companion.
neil@bradley.co.uk
www.bradley.co.uk