DTD validation

[techie]

I'm having a lot of trouble with a DTD that my xml is referencing. As soon
as I open the xml file in IE, I get the following error:
A name was started with an invalid character. Error processing resource
'file:///C:/BP/bp_testing/rtfml_doc.dtd'. Line 1, Position 1

#PCDATA | inline | character | formatting | link | plaintext | hiddentext |
a | xlink | linebreak | pagebreak | columnbreak | linktarget | fieldcode |
footnote | comment |
^
MSXML must have a problem with the '#' of #PCDATA that is defing in my dtd:

<!ENTITY % inline "#PCDATA | inline | character | formatting | link |
plaintext | hiddentext | a | xlink | %common;">
...
...
<!ELEMENT indexentry (level1 | level2 | level3 | level4 | level5 | level6 |
level7 | level8 | level9 | %inline;)*>

I don't know how to get round this error. The dtd is valid according to my
xml editor XML Spy but I have to make use of MSXML in order to transform the
xml file.

Any help would be appreciated.

[Richard Tobin]

In article <-b********************@biscit.net>,
techie <te****@nospam.biz> wrote:

<!ENTITY % inline "#PCDATA | inline | character | formatting | link |
plaintext | hiddentext | a | xlink | %common;"> <!ELEMENT indexentry (level1 | level2 | level3 | level4 | level5 | level6 |
level7 | level8 | level9 | %inline;)*>

#PCDATA must come first in the mixed context declaration. Move %inline;
to the start.

-- Richard

[techie]

"Richard Tobin" <ri*****@cogsci.ed.ac.uk> wrote in message
news:cj**********@pc-news.cogsci.ed.ac.uk...

In article <-b********************@biscit.net>,
techie <te****@nospam.biz> wrote:

<!ENTITY % inline "#PCDATA | inline | character | formatting | link |
plaintext | hiddentext | a | xlink | %common;">

<!ELEMENT indexentry (level1 | level2 | level3 | level4 | level5 | level6 |level7 | level8 | level9 | %inline;)*>

#PCDATA must come first in the mixed context declaration. Move %inline;
to the start.

Thanks. This seems to do the trick.