If you now try to process the xmllint'ed file in Sherlock (fetched via
just a normal HTTP GET and not via a cgi), can you now access any of
the nodes via XQuery?
On Monday, April 21, 2003, at 07:16 PM, Frangois Schiettecatte wrote:
Jussi-Pekka
I know that there is some junk in the file, well junk, Japanese
characters,
definitely not UTF-8.
I downloaded xmllint and it says nothing, so I assume that the XML is
ok. I
asked xmllint to reformat the code and it did it quite happily.
Cheers
Francois
On 4/21/03 9:34 PM, "Jussi-Pekka Mantere"
<email@hidden>
wrote:
Frangois,
Depending on what your server-side XML content is, SOAP::Lite 0.55
has
a bug where Unicode characters are munged. Validating the
on-the-wire
XML out-of-band (xmllint?) could yield some clues here.
Jussi-Pekka
On Monday, April 21, 2003, at 06:04 PM, Frangois Schiettecatte
wrote:
I am running into an interesting problem with the SOAP
implementation
in
Sherlock. It is croaking on some XML that is being returned from my
server.
Before people jump all over this and tell me this is my problem,
bear
with
me :) The XML in question is actually coming from a variety of
RSS/RDF
files. The thing that is interesting is that the SOAP client I have
developed for testing the SOAP server also croaks. The SOAP client
and
server are both developed using SOAPLite, a Perl module. What is
interesting
is that my SOAP client does not croak if I tell it _not_ to use the
expat
XML parser to process the XML returned from the SOAP server, but to
use
another (simpler) parser included for testing purposes.
