OK, I've checked this out. As I thought, this is problem of binary characters in the data.

Torgny, there is a character that looks like this in your XML (UTF 5 I think):

  <fontface></fontface>

It's in your DataEnvironment record (recno 45 in the FRX, objtype 25).

When I remove this character from the XML, the HTML generation process works fine (accented characters and all). There's nothing wrong with the XSLT, IOW.

Any idea how that character got there? If you have manually adjusted the report, then it is your responsibility not to do this <g>.

If you have not done this, however, we need to figure out what put it there.

When I edit the DE in a report, whether in 7, 8, or 9, the result is EMPTY(fontface) for this record, not a CHR(5). Do you have anything you can tell me about how you created this report that explains this?

FWIW the Tag information for that record does not look correct either. Below is the total contents (it looks like it leaves off mid-way in the Init method, notice no ENDIF).

I don't think *this* is an encoding issue (we encode the text properly, for example if you use < or > in your DE code you will see these characters as entity references in the resulting XML).

Can you let me know if this is actually what you have in your DE (it shouldn't be related, but perhaps it is)?

PROCEDURE Destroy
IF plMoreCopies = .F.
    RELEASE vp_EmbNum, vp_Artnr
ENDIF    

ENDPROC
PROCEDURE Init
#INCLUDE "INCLUDE\APPINCL.H"

LOCAL loDataEnvironment, ;
	  loAskForm 
	  
IF TYPE('vp_EmbNum') == "U"	  

  PUBLIC  vp_EmbNum, ;

>L<