> A question:
> Which features should be set by the setAtttribute() method other than the
> document handling?
I think we should use setAttribute for Xalan specific features, like
turning off structural rewrite optimization. I just want to avoid having a
bunch of parser specific features, and then having to maintain these.
-scott
Thomas2.Maesing
@bgs-ag.de To: [EMAIL PROTECTED]
cc: (bcc: Scott Boag/CAM/Lotus)
06/20/01 09:55 Subject: Antwort: Re: Antwort: Re:
Antwort: [Bug 2124] - Status of
AM SAXSource is changed in Xalan 2.1.0
Please respond
to xalan-dev
Yes,
it is right. But both Xpath Spec and XPointer Spec contain a warning that
the document must be valid to use this function. So the
user knows that he has to switch on validation. But the function can be
wrapped by other expressions which do not need a
valid document.
For some from SGML to XML converted documents there is no other chance
because such documents contain content models
which are not allowed in XML.
On the other hand XML Schema reactivates content models which were allowed
in SGML but not in XML DTDs.
There is another point. You can use a DTD for defaulting attributes and a
XML Schema for validating (e.g. The XML Schema for XML Schema
does it) . A not XML Schema supporting parser will fail then even if the
document is valid.
So to avoid user confusion dynamic validation should be switched off by
default until there is a common solution for such problems
( e.g. XSLT 2.0 , maybe JAXP 1.2 etc.).
A question:
Which features should be set by the setAtttribute() method other than the
document handling?
Regards
Thomas
[EMAIL PROTECTED] am 20.06.2001 14:42:53
Bitte antworten an [EMAIL PROTECTED]
An: [EMAIL PROTECTED]
Kopie:
Thema: Re: Antwort: Re: Antwort: [Bug 2124] - Status of SAXSource is
changed in Xalan 2.1.0
> If XSLT 2.0 becomes W3C Recommendation (or working draft) then it
> contingently makes sens to use dynamic validation as default because you
> need it to use the XMLSchema.
The id(...) function won't work correctly with validation turned off now,
correct?
-scott
Thomas2.Maesing
@bgs-ag.de To: [EMAIL PROTECTED]
cc: (bcc: Scott
Boag/CAM/Lotus)
06/20/2001 Subject: Antwort: Re:
Antwort: [Bug 2124] - Status of SAXSource is
04:00 AM changed in Xalan 2.1.0
Please respond
to xalan-dev
Yes,
I think so because it is the default behavior from xalan-j 1(in a later
version it was switched on, but as specified in the documentation of itself
it
should not. But this was no problem because errors did not stop the
transformation) up to xalan-j_2_1_0. It is also the default
behavior of Saxon.
I have found nothing in the XSLT Spec that a XSLT Processor should be a
validating XMLProcessor for the Source Document.
I also have found nothing in JAXP 1.1 Spec regarding to the transform
package.
So in my opinion at the time it should not be used as default.
If XSLT 2.0 becomes W3C Recommendation (or working draft) then it
contingently makes sens to use dynamic validation as default because you
need it to use the XMLSchema.
Regards
Thomas
[EMAIL PROTECTED] am 19.06.2001 19:18:36
Bitte antworten an [EMAIL PROTECTED]
An: [EMAIL PROTECTED]
Kopie:
Thema: Re: Antwort: [Bug 2124] - Status of SAXSource is changed in Xalan
2.1.0
Hi Thomas. I just want to be clear. You are proposing to turn off dynamic
validation altogether?
> Another possible way is to provide the direct validation/dynamic feature
> to the user
> as Attribute (via setAttribute(java.lang.String name, java.lang.Object
> value)) .
No. I don't want to get into the business of setting properties on the
XMLReader... that's what SAXSource is there for.
I think dynamic validation is the right default, but I understand your
problem. I really think this is a sort of Parser bug, but that's only my
opinion, and doesn't help much right now. I'll look into it a bit further.
-scott
