Framework can't read it's own XML files.
Mistake in IPDRXMLContentHandler.
Change following code to solve:
public void characters(char[] ch, int start, int end)
throws SAXException {
//EJM PATCH:
// It is illegal to access ch[0] if start is greater then
zero...
/*
String str = "";
if (ch[0] != '\n') {
str = new String(ch, start, end);
}
/**/
// Create empty string if current character block
starts with
// eof line marker. (Why?)
String str;
if (ch[start] != '\n') {
str = new String(ch, start, end);
} else {
str = "";
}
/**/
if (tagFlag_ && !(str.trim().equals(""))) {
appendTagValue_ = appendTagValue_ + str;
}
}
Logged In: YES
user_id=617954
The file is called IPDRXMLContentHandler.java to be more
precise
Logged In: YES
user_id=751148
Cannot reproduce. An example of an input XML file which
causes this problem would be helpful.
It would appear that any output created by the WriteTool is
properly read.
Logged In: YES
user_id=617954
The original code reads ch[0]. This is illegal if the value of
start is greater then 0. It seems this happens sometimes.
At the moment I'm not involved in IPDR related development,
and I have no XML file to reproduces available. Sorry about
that. Just read the docs about the "characters()" function.
(Btw, the solution is also mentioned in the first bug report, so
it shouldnt take to much time to fix)
Logged In: YES
user_id=751148
The testing cycle of the XML routines is done using the read
and write tools. These do not exhibit this behavior.
It does appear that an XML document which has whitespace
preceding the XMLDecl, will cause this behavior.
It turns out that it is invalid for an XML document to begin
this way. See the grammar for well formed documents in the
XML spec (section 2.8):
http://www.w3.org/TR/REC-xml
[22] prolog ::= XMLDecl? Misc* (doctypedecl Misc*)?
[23] XMLDecl ::= '<?xml' VersionInfo EncodingDecl?
SDDecl? S? '?>'
A more meaningful exception will be thrown in subsequent
release.