- From: Robert Miner <RobertM@dessci.com>
- Date: Tue, 11 Jan 2005 12:54:32 -0600
- To: Bernhard.Keil@soft4science.com
- CC: davidc@nag.co.uk, www-math@w3.org
Hi Bernhard, You wrote: > Its definitely not the source file, as the same error occurs with this minimal XHTML+MathML document: > > <?xml version="1.0" encoding="UTF-8"?> > <?xml-stylesheet type="text/xsl" href="http://www.w3.org/Math/XSL/pmathml.xsl"?> > <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1 plus MathML 2.0//EN" "http://www.w3.org/Math/DTD/mathml2/xhtml-math11-f.dtd"> > <html xmlns="http://www.w3.org/1999/xhtml"> > <head> > <title></title> > </head> > <body> > </body> > </html> > > I dont understand it, as it worked before since at least 12 month > and it still workes on my other computer (except that this error occurs very, very seldom). > > Nevertheless, I might try anaylizing the stylesheet and the DTD as you suggested. Did you perhaps just install Office 2003 on the problem machine? We discovered that Office 2003 installs a filter that sniffs text/xml documents, and presumably does something to Office XML documents. We found out about it since it overwrites MathPlayer's filter, which attempts to sniff out XHTML+MathML documents. MathPlayer is not supposed to filter any document with a stylesheet PI, so I doubt it is the problem, but I guess the MathPlayer filter could have some bug too. Anyway, if a filter does have a bug affecting plane 1 characters, then by the time the XSL tranform tries to run, the source has already been corrupted. The corruption would only be in the copy in memory -- view source, etc wouldn't show it. The fact that the bad characters from the error message don't appear in the source, stylesheet or DTD make it sound as if some piece of software is corrupting the document in memory, which lead me to think of filters. To eliminate MIME filters as a possible cause of the problem, check your registry. The MIME filters are listed in HKEY_CLASSES_ROOT\PROTOCOLS\Filter You can turn off any text/xml filtering by deleting the text/xml subkey (after saving it, naturally, in case you want to put it back...) --Robert ------------------------------------------------------------------ Dr. Robert Miner RobertM@dessci.com W3C Math Interest Group Co-Chair 651-223-2883 Design Science, Inc. "How Science Communicates" www.dessci.com ------------------------------------------------------------------
Received on Tuesday, 11 January 2005 18:55:07 UTC