[reportlab-users] RE: pyRXP can't handle simplified docbook dtd

[Andy Robinson]

Please note that our users list address changed
away from Yahoo recently; the correct one is
listed above.

pyRXP does not handle wide characters as currently 
compiled.  That's because Python already has good unicode
libraries and our philosophy was for the programmer
to use these and take charge of any encoding conversions.
You can convert to UTF8 with one line of Python beforehand.
If you put UTF8 in, you'll get UTF8 out, and so on.

You CAN compile the sources with wide character support
but we didn't.  If you need to do this, my colleague
Robin Becker (robin@reportlab.com) can help you.

- Andy Robinson



> -----Original Message-----
> From: tympanum2001 [mailto:mats@svensson.ms]
> Sent: 18 June 2002 08:51
> To: reportlab-users@yahoogroups.com
> Subject: pyRXP can't handle simplified docbook dtd
> 
> 
> Windows 2000
> ActiveState Python 2.2
> pyRXP.version is 0.7
> I use CR2 of the simplified docbook:
> http://www.oasis-open.org/docbook/specs/wd-docbook-simple-1.0-CR2.html
> 
> If I try to parse a doc and validate it against the 
> flattened sdocbook.dtd I get lots of complaints about invalid 
> character entities. 
> 
> Are the precompiled windows binaries unable to handle wide 
> characters or is it some other problem?
> 
> Traceback:
> pyRXP.Error: Error: 0x215b is not a valid 8-bit XML character
>  in unnamed entity at line 320 char 26 of file:///C:/xml/dtd/sdocbook.dtd
> 
> This is line 320:
> <!ENTITY frac18	"&#x215B;">
> 
> Any suggestions? 
> 
> /Mats
> 
>