On Wed, Sep 13, 2006 at 01:10:20PM -0700, Wendell Nichols wrote: > Thankyou gentlemen for you're interesting ideas. > What I've done today is: > > * recompiled the libxml2 code with preparser option CONV(UTF-8) to > keep the embedded constants as ascii/utf-8 chars.
okay, > * Because this breaks the IO code (file open mode strings like > "rb" no longer ebcdic), I frontended the samples with some code to read Well if you have specific zos patches for those parts that can be isolated cleanly (for exemple with #ifdef ZOS or another symbol) then submit them so we can keep an unified code. > the xml documents into a char buffer. (I didn't experiment with not > using the conv option on the io files because mixing the encodings would > have gotten confusing... > * Then I parsed a utf-8 encoded document successfully! > > I copied the group so that this info would be available to the community > at large. Thanks, but a patch correcting the problem directly at the code level would be more convenient if someone else ever decide to run libxml2 in your environment. Daniel -- Red Hat Virtualization group http://redhat.com/virtualization/ Daniel Veillard | virtualization library http://libvirt.org/ [EMAIL PROTECTED] | libxml GNOME XML XSLT toolkit http://xmlsoft.org/ http://veillard.com/ | Rpmfind RPM search engine http://rpmfind.net/ _______________________________________________ xml mailing list, project page http://xmlsoft.org/ [email protected] http://mail.gnome.org/mailman/listinfo/xml
