XMLPPM 0.95 BUGS

James Cheney 

11/24/2004

0.98.2 incorporates a serious bug fix.  Because of an off-by-one error,
the encoding was broken for large files.

6/7/2004

Release 0.98 includes some bug fixes:

1.  A bug involving namespace attributes.  They are now treated
identically to other attributes.

2.  A coding problem that gcc 3.4 is picker about than earlier gccs.

3.  In the previous version, decompression was substantially slower
because internationalized files (IFiles) were not buffered.  This has
been fixed.

3/19/2001

As far as I know, XMLPPM might work on all XML documents.  I have tested
it on a wide variety of XML documents, and found and fixed many bugs.
It's likely that there are still some in there.

XMLPPM doesn't compress the XML text directly, but rather the SAX events
generated by expat as it parses.  This makes XMLPPM slightly lossy
in that some information such as formatting inside DTDs is not reported in
the SAX events.

Also, XMLPPM runs into problems with entities.  Currently, XMLPPM 
conservatively replaces all occurrences of reserved characters such as 
&, ;, and < with their predefined entity references.  This may change 
your document in an essential way.  Be warned!

5/19/2001

Found and fixed a bug that occurs when an XML file begins with a comment
or processing instruction.

Currently, XMLPPM is limited to around 250 element and attribute symbols.
If this is exceeded, XMLPPM will crash.

Currently, XMLPPM has a hard-wired memory limit of 10MB.  If this is exceeded,
the program crashes.

2/11/2003 

Replaced the old PPM encoder with a new one based on Dmitry Shkarin's
ppmdi.  It is much faster and better, and also manages memory safely,
so that if the model runs out of memory then we delete it and start a
new one.

I added support for handling start/end namespace SAX events, but haven't
tested it thoroughly.