Does the current schema support expose the type information to the
client code, or is it purely used for validation, e.g. given a schema
that defines an element 'fred' to be of type xsd:string, and given an
instance doc of
http://www.w3.org/2001/XMLSchema-instance";
xsi:noNamespaceSchemaLocation
Andy,
Is it possible to share ICU built for XML and ICU used by other
applications?
Carl
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]]On Behalf Of Andy Heninger
> Sent: Tuesday, July 03, 2001 2:15 PM
> To: ICU Mailing List
> Cc: [EMAIL PROTECTED]
> Subject:
> Andy,
>
> Is it possible to share ICU built for XML and ICU used by other
> applications?
I was going to ask something similar. I submitted a patch for Xerces a while
ago that let one do just that. The reason was that I am packaging ICU for
Debian GNU/Linux (watch for the 1.8.1 packages soon,
Tinny,
Sorry to be dense, but how does this perl script resolve the
change in ICU's data filename? Does it actually just rename
the file? I hope not, as the name was changed to solve a
problem with collisions with other versions of the package.
But I'm not a perl maven, and I couldn't figure it
Hi
I have a question
about how to selectively disable parsing of certain tags? The situation is
I have a html file that displays well in browser but not conform to the xHTML
standard, can I still use xerces-c to do some parsing, say i
have:
my page
http://www.someplace.com"
Andy Heninger wrote:
>
> Are you using the "packageBinaries.pl" script from Xerces to do the build?
> There's a fair amount of twiddling that happens to switch the xerces build
> from stand-alone mode to the use of ICU, and this script automates it. It
> is also what the xerces group uses to do
What about
delete __[ ]__ str;
Regards,
Peiyong Zhang
XML Parsers Development
IBM Toronto Laboratory email: [EMAIL PROTECTED]
Phone: (416)448-4088; Fax: (416)448-4414; T/L: 778-4088
[EMAIL PROTECTED] on 07/03/2001 04:29:39 PM
Please respo
"Tinny Ng" <[EMAIL PROTECTED]> writes:
> If it works for most other IDOM methods, except
> getElementsByTagName(), then there may be a bug in this function
It's not just a linux issue, I replicated it under Solaris:
$ uname -a
SunOS usf-cf-sparc-solaris-2 5.8 Generic_108528-05 sun4u sparc SUNW,
The calls look OK, but each one is a memory leak, since you are required to
delete the pointer returned from XMLString::transcode():
XMLCh* str = XMLString::transcode("foo");
delete [] str;
Dave
"Tinny Ng" <[EMAIL PROTECTED]> writes:
> If it works for most other IDOM methods, except getElementsByTagName(),
> then there may be a bug in this function
>
> Can you post your modified sample program for further investigation?
Here's the version that segfaults using getElementsByTagName(
[EMAIL PROTECTED] writes:
> + Then, I'm getting pointer to a newly allocated buffer of char elements:
> + char* str = value.transcode();
> +
> + Then, I'm going to delete that allocated area:
> + delete str; // DAMAGE before free block
> +
> + What do I do wrong? I was looking into your code in
Are you using the "packageBinaries.pl" script from Xerces to do the build?
There's a fair amount of twiddling that happens to switch the xerces build
from stand-alone mode to the use of ICU, and this script automates it. It
is also what the xerces group uses to do its full builds of XML4C. In
X
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=2440
*** shadow/2440 Tue Jul 3 13:29:39 2001
--- shadow/2440.tmp.11105 Tue Jul 3 14:07:03 2001
***
*** 2,9
| delete on pointer from transcode() failed |
+---
"Tinny Ng" <[EMAIL PROTECTED]> writes:
> If it works for most other IDOM methods, except getElementsByTagName(),
> then there may be a bug in this function
>
> Can you post your modified sample program for further investigation?
Hey Tinny,
Here's some more info. The following patch which
Try using the supplied script packageBinaries.pl to build the library, this script
will fix the missing link for you.
set ICUROOT=
set XERCESCROOT=
perl packageBinaries.pl -s -o -t icu
Tinny
Jack Marr wrote:
> I'm trying to build XML4C 3.5.0 from sources provided by
> http://www.a
"Tinny Ng" <[EMAIL PROTECTED]> writes:
> Yes this is on purpose. As you may know the DOM Level 3 exposes the
> information contained in the declaration as attributes of the
> Document (and Entity). XMLDecl node is non-standard compliant, and
> we purposely didn't want to encourage that in the n
"Tinny Ng" <[EMAIL PROTECTED]> writes:
> If it works for most other IDOM methods, except
> getElementsByTagName(), then there may be a bug in this function
>
> Can you post your modified sample program for further investigation?
Hey Tinny,
Thanks for the response.
Here's the CVS diff output
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=2440
*** shadow/2440 Tue Jul 3 13:29:39 2001
--- shadow/2440.tmp.10932 Tue Jul 3 13:29:39 2001
***
*** 0
--- 1,37
+ ++
+ | delete on pointer
I'm trying to build XML4C 3.5.0 from sources provided by
http://www.alphaworks.ibm.com/tech/xml4c, but am stymied at the point where it
wants to link Xerces with ICU because it is looking for libicudata.so, which is
not produced by ICU, as far as I can tell. It does produce libicudt18b.so and
and
Found 'std' used in XMLDeclImpl.cpp and DOMParser.cpp which are now fixed
in CVS.
Tinny
Dean Roddey wrote:
I'm
not monitoring the list anymore but here's one that I ran into at work,
using the HP compiler (which is evidentally more anal about this than most
others.) In the two
classes in DOM th
Aniruddha,
> ... the DOM_NodeList.hpp has the following files included within itself ...
The DOM_NodeList.hpp shipped in Xerces 1.4 has the following lines:
#ifndef DOM_NodeList_HEADER_GUARD_
#define DOM_NodeList_HEADER_GUARD_
#include
#include
class NodeListImpl;
Hi ,
I am trying to write the following code but i am getting errors.the
error is pasted below
for (int i=0 ; i < var.getLength () ; i++)
{ DOM_Node temp = var.item(i);
DOM_CharacterData text = (DOM_CharacterData)temp.getFirstChild();
}
tilde/lpcoop/xmlparse
Column number is counted in chars.
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Michael,
Although there are a few hundred of @STO, there should be only one @ST01985
Anyway are you using Iconv Transcoder? If so, there is a known bug which may
explain your link problem (Bugzilla bug #2393
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=2393).
While waiting for the fi
Houamn,
One suggestion is to set "reuseGrammar" to true during second xml document
parse. It will reuse the grammar saved in the first xml document parse, and
ignore whatever grammar specified in the second xml document. Note you cannot
have internal subset in this case.
Another suggestion is
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=2391
*** shadow/2391 Thu Jun 28 20:24:30 2001
--- shadow/2391.tmp.9043Tue Jul 3 10:30:31 2001
***
*** 2,9
| Problem in using Parser in solaris |
+---
Jason,
If it works for most other IDOM methods, except getElementsByTagName(),
then there may be a bug in this function
Can you post your modified sample program for further investigation?
Thanks!
Tinny
"Jason E. Stewart" wrote:
> "Jason E. Stewart" <[EMAIL PROTECTED]> writes:
>
> > My
Jason,
> Does anyone care about keeping the docs up-to-date?
We care. But in reality, it always got over-looked and not treated as
high priority.
> If so, should I file bugs when I find documentation problems like this?
Sure, and if you can provide us information what missing exactly (or even
Yes this is on purpose. As you may know the DOM Level 3 exposes the
information contained in the declaration as attributes of the Document
(and Entity). XMLDecl node is non-standard compliant, and we purposely
didn't want to encourage that in the new IDOM.
Tinny
"Jason E. Stewart" wrote:
> Th
This may be related to Bug #812 in Bugzilla
(http://nagoya.apache.org/bugzilla/show_bug.cgi?id=812) which is now fixed in
the latest nightly build http://xml.apache.org/dist/xerces-c/nightly/.
Tinny
Andrew Snare wrote:
> [apologies if this is a duplicate, my original submission seems to have
>
Just applied. Would you please verify?
Thanks!
Tinny
Miroslaw Dobrzanski-Neumann wrote:
> Dear all,
>
> attached you find a patch supporting dce threading on:
> - AIX 4x with IBM xlC_r4 compiler
> - SunOS with SUN CC compiler and Transarc DCE implementation (transarc belongs
> to IBM); it s
Just applied. Would you please verify?
Thanks!
Tinny
Miroslaw Dobrzanski-Neumann wrote:
> On Sun, Jul 01, 2001 at 06:50:02AM -0700, James Berry wrote:
> > Carl,
> >
> > You've got some great suggestions...how about submitting some great patches
> > to support them? ;)
>
> A week ago I have su
I need to know the real position where xml declaration is ended.
I use my own class which should resolve this task (see the pseudo code
below)
class XmlInfoParser : public SAXParser, public HandlerBase
{
public:
XmlInfoParser();
virtual void XMLDecl(
const XMLCh* const versionSt
Source: XMLScanner.cpp
XML prolog : chUnicodeMarker +
curReader->getColumnNumber() returned 8, so the following code failed and
error was emitted.
Please see your code, whether you consider the Unicode marker when
determining the column number.
if (checkXMLDecl(true))
You've got a number of choices. First, don't call Terminate(). It's there to
release memory that's lazily allocated; it'll be released automatically when
your process exits.
Second, call Initialize() when your process fires up, without a matching
Terminate(). This has the same effect as the first
I need to get the line and the column from the file
parsed.
So I have done the folowing implementation
:
The "X.h" :
class ReaderMgr;
class X: public HandlerBase
{public: X(); virtual ~X();
void startDocument(); void
endDocument();
void startElement(const XMLCh* const
hello,
i am using a xmlparser both saxparser and domparser, both working in a
Single Process, Saxparser will be invoked by some event , similarly
Domparser by some other event, my problem is that i want to call
XMLPlatform::initialize (), more than once in a process, something like this
XMLPlatfo
Hi Curt,
I tried loadXML("") and I get the same exception! so I guess it is a
problem with Xerces-C, I tried all dll versions from 1.2 till 1.5 and it
still works with 1.3 but not anymore with 1.4 or 1.5.
Andre
From: "Curt Arnold" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
> >I get the follo
38 matches
Mail list logo
