<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta http-equiv="Content-Type" content="text/html;charset=ISO-8859-1">
<title></title>
</head>
<body>
A half an hour digging in the logs and I think I have an explanation:<br>
0) The xmldsigverifier was compiled in April 2002 and it is more
than<br>
a year old now (probably I need to upgrade it :) )<br>
1) The c14n code in libxml2 version 2.4.20 that was used to
compiled <br>
xmldsigverifier returns the exact results as you describe<br>
2) The namespace processing in c14n.c was fixed around July 31, 2002<br>
in order to support a new Merlin's c14n tests (merlin-c14n-three).
As far as <br>
I can remember and as far as I can see from the code, this changes<br>
solves exactly this problem.<br>
<br>
Bottom line: there was a bug and it was fixed almost a year ago,
xmldsigverifier<br>
on the web site is obsolete (and I hope I will have time to update it
soon).<br>
<br>
Now I would like to repeat my explanations. I would appreciate if Rich
or<br>
someone else familiar with c14n sepcifications:<br>
<br>
We have something like this: <br>
<Root xmlns=<a class="moz-txt-link-rfc2396E"
href="http://examples.com">"http://examples.com"</a>> <br>
<Object>Test</Object> <br>
</Root> <br>
According to the spec [1] , the non-default namespace node is <br>
rendered only if it is in the XPath node-set. In our case, <br>
the XPath expression selects <b class="moz-txt-star"><span
class="moz-txt-tag">*</span>only<span class="moz-txt-tag">*</span></b>
<Object/> node itself <br>
and none of its namespaces or attributes nodes. Thus I think <br>
that xmlsec/libxml do the right thing by returning <br>
<Object></Object> <br>
after c14n.<br>
<br>
<br>
Aleksey<br>
<br>
<br>
[1] <a class="moz-txt-link-freetext"
href="http://www.w3.org/TR/2001/REC-xml-c14n-20010315#ProcessingModel">http://www.w3.org/TR/2001/REC-xml-c14n-20010315#ProcessingModel</a><br>
<br>
</body>
</html>