XML and special characters ...

[Alan Searle]

I generate XML from an MS-Access 2002 database and find that I can
format and display the data no problem with a XSL/HTML template. Perfect!

However, then I found that I needed to replace my automatic MS-Access
output with a customised dump of the XML and find that I am now having a
problem with special characters (umlauts).

For my tests I have built two identical extracts one generated by
MS-Access and the other constructed by me using VBA. Here I find that
my (VBA generated) version produces an error message 'illegal
characters' (stopping at the point where umlauts: ä and ü are found) in
the browser while the MS-Access generated version still displays.

It can't be anthing in the content because both files are identical so I
am wondering whether I need to make any specific declarations when I
open the text file in VBA? Yes, are there any hidden commands which
indicate which characterset the file will contain?

Any tips or links on the issue of handling special characters in XML
would be a great help.

Many thanks,
Alan Searle

[Steve]

I don't know about a problem with umlauts and Access specifically, but
I do know that XML requires many special characters to use encoded
equivalents. For instance, the & character has to be &amp; , < has to
be &gt; , etc.

I use an ASP.NET page to write text to XML files (they drive my
website) and I've found that ASP.NET automatically converts special
characters to their XML equivalents. If nothing else, you may consider
using an external ASP.NET page to create your XML files from your
Access tables.

[David W. Fenton]

"Steve" <th*********@gmail.com> wrote in
news:11**********************@o13g2000cwo.googlegr oups.com:

I don't know about a problem with umlauts and Access specifically,
but I do know that XML requires many special characters to use
encoded equivalents. For instance, the & character has to be &amp;
, < has to be &gt; , etc.

Actually, it would depend on which DTD you're using. I don't do XML
(or XHTML), but I do know that with the HTML transitional DTD, you
don't have to encode the characters, but with the strict DTD, you
*do* have to encode them.

I have a table and a couple of functions for HTML entity encoding.
I'd be willing to post them on my website if anyone is interested.

But if the content is identical in the output (i.e., the one the
works doesn't entity encode the umlauts), then it must just be a DTD
declaration that needs to be changed, and you wouldn't need to worry
about encoding the data. Of course, you must look at the XML in a
plain-text editor to see if the encoding is there or not, as a
browser won't show you the encoding, just the end result (unless you
view source).

--
David W. Fenton http://www.bway.net/~dfenton
dfenton at bway dot net http://www.bway.net/~dfassoc