On 18 Dec 2010, at 10:19, Niels Grewe wrote: > On Fri, Dec 17, 2010 at 01:31:19PM +0000, Richard Frith-Macdonald wrote: >> I just want to remove the dependency on libxml2 ... I was under the >> impression that the built-in parser worked for all valid XML. >> We want the built-in parser anyway, in order to parse apple property lists >> because they contain escape characters which are rejected by libxml2 ... >> (the builtin/sloppy parser accepts escapes as if they were legal). > > Sounds reasonable. > >> If you can produce a testcase demonstrating what's wrong/different I will >> try to code a fix (and in the meantime you can easily revert to use libxml2 >> locally of course). > > Okay. I'll put something together (I think I need to check back with my > MacOS machine for the actual behaviour of Apple's NSXMLParser). Two > problems I already could identify are the following: > > * Unless you call -_setAcceptHTML: with YES, any XML data without > <?xml…?> prolog is rejected by the parser, although it's perfectly > well-formed (according to §2.8 of the XML spec). > * It generates a -parser:didStartElement:… call to the delegate for the > doctype declaration, which it probably shouldn't because it only > specfies the grammar to use for validating the XML.
Thanks ... I guess the thing to do is create testcases and add them to testsuite in svn to see how these things behave. That way we can easily check that the GNUstep and OSX behaviors are the same simply by running the testsuite on OSX with the Apple Foundation, and on a GNUstep system. I guess it might be the case that the GNUstep libxml2 implementation of NSXMLParser is actually 'wrong'... if it differs from what Apple's NSXMLParser does. _______________________________________________ Gnustep-dev mailing list Gnustep-dev@gnu.org http://lists.gnu.org/mailman/listinfo/gnustep-dev