libxmlplusplus-general — Open discusions about libXML++

[Libxmlplusplus-general] Bug in SaxParser::parse_stream?

[Fredrik A. <e97_far@e.kth.se>]

Hello,

I think I have found a bug in SaxParser::parse_stream (saxparser.cc)
in libxml++ 0.17.0. This code:

  if( ! _exception )
  {
    exception * tmp = _exception;
    _exception = 0;
    tmp->Raise();
  }
  
will either do nothing at all or try to follow a null pointer.

I have attached a patch for what I think was intended.

/Fredrik Arnerup
e97_far@e.kth.se

[Libxmlplusplus-general] Re: newbie question: libxml++ vs. xerces-C++

[Greg S. <gr...@ho...>]

Just to make sure... 

When you talk about switching to Glib::ustring in libxml++, you 
*are* planning to do it in such a way that one can still use 
std::string if they require no unicode support, correct?  

Thanks,

Greg

Re: [Libxmlplusplus-general] multi-node-types API changes

[Christophe de V. <cde...@al...>]

=2D----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Mardi 17 D=E9cembre 2002 22:25, Christophe de Vienne wrote:
>  - Wouldn't a Clone() method simplify the Node(const Node * from)
> implementation ?

It should even replace it. We cannot leave such a constructor on an abstact=
=20
class.
=2D----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (GNU/Linux)

iD8DBQE9/5oFB+sU3TyOQjARAmZsAJ4lRKHW08mbeG2d6k60r+tLV+A+TACcCH6w
dpHY04zUxKUA22z1R2R0nd4=3D
=3DDqlP
=2D----END PGP SIGNATURE-----

[Libxmlplusplus-general] multi-node-types API changes

[Christophe de V. <cde...@al...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Murray,

Having a quick look at what you commited, I have a few small remarks, beside 
the fact that what you've done correspond almost exactly with what I was 
thinking about (yeepee we do agree ;-)

 - The TextNode doesn't need all these constuctors since it has no name (well, 
always the same), and no line number in libxml, although we could 
automaticaly give the parent line number to it.
 - As I said a couple of week ago, I do agree with you on the problem caused 
by the methods returning const references. So the name() method can become
string name() const;
 - Wouldn't a Clone() method simplify the Node(const Node * from) 
implementation ?
 - don't you think we could get rid of the _initialised attribute ? The only 
case it is set to false is when the name of the node is empty. Is it really 
necessary ? I have the feeling that not. If something wrong happens during a 
node initialisation, then we throw an exception...


As far as the children API in the Node type, I guess that keeping it there 
simplify it's use : there is not need to make any difference between all 
types of node to go through a tree. A TextNode is a node with 0 children 
after all.
So let's keep it like this as you suggest.



-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (GNU/Linux)

iD8DBQE9/5Y5B+sU3TyOQjARAnGzAJsFtEaNJ+8w0fq2GZ6PqFSR25Kc8wCgvZS1
5CIuQWK3FtVNddgfg18OHZ4=
=23g5
-----END PGP SIGNATURE-----

Re: [Libxmlplusplus-general] newbie question: libxml++ vs. xerces-C++

[Christophe de V. <cde...@al...>]

=2D----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Le Lundi 16 D=E9cembre 2002 20:12, Ed Hill a =E9crit :
> Hi folks,
>
> I've started converting to XML for some of my data sets (mostly X-ray
> spectra and geometry info for GPL'd scientific computing applications)
> and I was wondering if someone could provide some comments on the
> ups/downs the Xerces-C++ (http://xml.apache.org/xerces-c/index.html)
> parser versus the libxml++ parser.  I've written some demo code using
> both and, so far, I prefer the cleaner, simpler libxml++ SAX interface.
> I find the Xerces code to be a lot more confusing...

If you're looking for simplicity of use, libxml++ may be better. If you're=
=20
looking for a more complete XML parser, then Xerces is probably the best of=
=20
both.

>
> Specific questions:
>
>   1) Does (or can) libxml++ do validation while parsing?  I'm writing
>      DTDs for my data sets and would prefer a validating parser.
>

No. But I want to add DTD validation support in libxml++ in the future (I h=
ave=20
no precise delay right now, since there is more urgent things to do).

>   2) Xerces-C++ seems to use Unicode while libxml doesn't.  Should I
>      care?  Or am I missing something?

libxml does, but libxml++ does not. This is indeed one limitation of libxml=
++,=20
which should be solved one day, but we have no precise plan about this for=
=20
now.



Christophe
=2D----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iEYEARECAAYFAj3+6PUACgkQB+sU3TyOQjBcywCfTnn9A240k3PBAK8bqiS80TiT
pu8An1OIiRgTDZMFYjgZ/PU/Jo9vH+Ta
=3DRVoF
=2D----END PGP SIGNATURE-----

[Libxmlplusplus-general] newbie question: libxml++ vs. xerces-C++

[Ed H. <ed...@eh...>]

Hi folks,

I've started converting to XML for some of my data sets (mostly X-ray
spectra and geometry info for GPL'd scientific computing applications)
and I was wondering if someone could provide some comments on the
ups/downs the Xerces-C++ (http://xml.apache.org/xerces-c/index.html)
parser versus the libxml++ parser.  I've written some demo code using
both and, so far, I prefer the cleaner, simpler libxml++ SAX interface.=20
I find the Xerces code to be a lot more confusing...

Specific questions:

  1) Does (or can) libxml++ do validation while parsing?  I'm writing=20
     DTDs for my data sets and would prefer a validating parser.

  2) Xerces-C++ seems to use Unicode while libxml doesn't.  Should I=20
     care?  Or am I missing something?

Any help is appreciated!

thanks,
Ed


--=20
Edward H. Hill III, PhD=20
Post-Doctoral Researcher   |  Email:  ed...@eh...,  eh...@mi...
Division of ESE            |  URLs:   http://www.eh3.com
Colorado School of Mines   |    http://cesep.mines.edu/people/hill.htm
Golden, CO  80401          |  Phones:  303-384-2094, 303-273-3483

Re: [Libxmlplusplus-general] Development Status

[Christophe de V. <cde...@al...>]

=2D----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Lundi 16 D=E9cembre 2002 19:36, darco wrote:
> I just thought that I should point out that having the "Development
> Status" for this project on the sourceforge page being set to "5 -
> Production/Stable" is misleading.
>

yes

> Shouldn't the Development Status be set to 2 (Pre-Alpha) or 3?
>

yes too...

I'll update the pages ASAP

> --- darco
> http://www.deepdarc.com/
>
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by:
> With Great Power, Comes Great Responsibility
> Learn to use your power at OSDN's High Performance Computing Channel
> http://hpc.devchannel.org/
> _______________________________________________
> Libxmlplusplus-general mailing list
> Lib...@li...
> https://lists.sourceforge.net/lists/listinfo/libxmlplusplus-general
=2D----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (GNU/Linux)

iD8DBQE9/jRqB+sU3TyOQjARAjQkAKDYMUdxMWSNg2vsW6PEC9ojy/3m2ACgsfxJ
gIdk0qefM/+f+RYIPZ2fPtA=3D
=3DkJ+d
=2D----END PGP SIGNATURE-----

[Libxmlplusplus-general] Development Status

[darco <da...@de...>]

I just thought that I should point out that having the "Development 
Status" for this project on the sourceforge page being set to "5 - 
Production/Stable" is misleading.

Shouldn't the Development Status be set to 2 (Pre-Alpha) or 3?

--- darco
http://www.deepdarc.com/

Re: [Libxmlplusplus-general] write() broken, maybe fixed

[Christophe de V. <cde...@al...>]

=2D----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Le Lundi 16 D=E9cembre 2002 16:09, Christophe de VIENNE a =E9crit :
> > Until we have some concept of text nodes I don't think we should demand
> > that people know what is a text node.
>
> agree.

Just a precision : In my opinion if we consider that people use add_content=
()=20
( and not set_content() ) and the get_content() accessor, they don't need t=
o=20
know what's underlying.

What I'll do on the fix release is to document the content accessors to avo=
id=20
confusion.

=2D----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iEYEARECAAYFAj397tAACgkQB+sU3TyOQjDDtgCggeTUnlGXSaZdGdNswmSWVL0X
i/AAn2zzJ/VZx0PCf/FAZ4Iv3aN+WjzW
=3D3x1k
=2D----END PGP SIGNATURE-----

[Libxmlplusplus-general] SAXParser: preventing expansion of predefined entities

[Jonathan W. <co...@co...>]

Hi guys,

I've been playing around with libxml++ for a couple of days, trying to
figure out if we want to use it in a new project we're starting. It
looks like we'll be using libxml, but would like a nice C++ interface
to it.

AFAICT there's no way to use the libxml++'s SAX interface without
expanding the standard entities (& etc), is that right?

To clarify, I'm parsing an XML doc with the SAX parser, outputting it
with some extra elements added dynamically (this is only a test for now,
so ignore whether or not this is the best technique :-)
I then pass the resulting output to libxslt. The libxml++ SAXparser's
_callback_get_entity() member is private, so I can't override it, and
it always expands "&" to "&" which causes an error when the doc is
passed to XSLT.

This behaviour ignores the value of the xmlSubstituteEntitiesDefault
global property, and AFAICT there's no way to change it (if I'm wrong
here, please point me the right way and I'll stroll off in that
direction!)

Is there a good reason for this, or would you be willing to accept a
patch to (off the top of my head) make _callback_get_entity() call a
protected virtual function, so that derived classes can alter the
behaviour?

Thanks for your time,

jon


-- 
"There are basically two types of people. People who accomplish things,
 and people who claim to have accomplished things.
 The first group is less crowded."
	- Mark Twain

[Libxmlplusplus-general] libxml++-0.17 release

[Christophe de V. <cde...@al...>]

Hi all,

The 0.17 version of libxml++ has just been released.
You can download it from sourceforge by the folowin link :
http://sourceforge.net/project/showfiles.php?group_id=12999&release_id=127313

Noticable changes since 0.16 version are :
* properties are now called attributes.
* it is now possible to throw exceptions from callback
methods in SaxParser.
* parsers can parse from a stream.
* memory leak fixed in SaxParser.
* configure check for libxml2 2.4.1 at least.
* bug fixes.

For more details, please see Changelog in the distribution package.

Have fun ;-)

Christophe

Re: [Libxmlplusplus-general] Creating Comment Nodes

[Christophe de V. <cde...@al...>]

=2D----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Samedi 7 D=E9cembre 2002 01:48, darco wrote:
> Is there a way to create a comment node? For my save procedure, I
> wanted to add a comment to the file that stated the time it was saved,
> who saved it, etc... I see that I can parse comment nodes, and that
> they have a name of "comment". However, when I try to create such a
> node in reverse, it fails in the obvious way.

This type of node is not yet supported in libxml++.
Since I wanted to clarify the use of content nodes, I'll add this at the sa=
me=20
time.
Expect something be the end of the next week.

Cheers,

Christophe

>
> --- darco
> http://www.deepdarc.com/
>
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Libxmlplusplus-general mailing list
> Lib...@li...
> https://lists.sourceforge.net/lists/listinfo/libxmlplusplus-general
=2D----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (GNU/Linux)

iD8DBQE98eNrB+sU3TyOQjARAqC4AJ0bdRbjD6gHU3j05YyO4KWfZVIViQCbBbaS
UGSiLpBdHb7LCXL0pwp9nTM=3D
=3D5aZn
=2D----END PGP SIGNATURE-----

[Libxmlplusplus-general] Creating Comment Nodes

[darco <da...@de...>]

Is there a way to create a comment node? For my save procedure, I 
wanted to add a comment to the file that stated the time it was saved, 
who saved it, etc... I see that I can parse comment nodes, and that 
they have a name of "comment". However, when I try to create such a 
node in reverse, it fails in the obvious way.

--- darco
http://www.deepdarc.com/

Re: [Libxmlplusplus-general] xmlpp::Node::line()

[Christophe de V. <cde...@al...>]

=2D----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

oops, sorry. Having a look on the code, you're right. In the rewritting of =
the=20
parsers the option has disapeared.
This should be fixed in the CVS tomorrow at last.

Cheers,

Christophe

On Jeudi 5 D=E9cembre 2002 21:55, Christophe de Vienne wrote:
> Hi,
>
> On Jeudi 5 D=E9cembre 2002 21:43, darco wrote:
> > However, I notice that xmlpp::Node::line() seems to only return zero.
>
> The content nodes always have their line number to zero.
> Try to get the line number of the parent node which should give you the
> something correct.
>
> Cheers,
>
> Christophe
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
>
> Libxmlplusplus-general mailing list
> Lib...@li...
> https://lists.sourceforge.net/lists/listinfo/libxmlplusplus-general
=2D----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (GNU/Linux)

iD8DBQE9774qB+sU3TyOQjARAl1OAKDprygTr8S1ur9DTUJaqJHFJ8dWswCfSqWM
=46q9tBPSYG4ht2cMhA6g432I=3D
=3DOeT4
=2D----END PGP SIGNATURE-----

Re: [Libxmlplusplus-general] xmlpp::Node::line()

[Christophe de V. <cde...@al...>]

Hi,

On Jeudi 5 D=E9cembre 2002 21:43, darco wrote:
> However, I notice that xmlpp::Node::line() seems to only return zero.

The content nodes always have their line number to zero.
Try to get the line number of the parent node which should give you the=20
something correct.

Cheers,

Christophe

[Libxmlplusplus-general] xmlpp::Node::line()

[darco <da...@de...>]

I'm writing a XML DOM parser for a project that I am working on. When I 
was reading over the API for libxml++, I saw the member function 
xmlpp::Node::line(), and assumed that it returned the line number that 
the node started on. So, I used it in my error reporting functions to 
indicate which lines contained errors... (Errors in content, not 
well-formedness)

However, I notice that xmlpp::Node::line() seems to only return zero.

Did I get the wrong idea about this member function? Is there some step 
that I'm missing? Or is this perhaps a bug...?

- darco
http://www.deepdarc.com/

[Libxmlplusplus-general] has_content()

[Christophe de V. <cde...@al...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi murray,

I've just realised that in node.cc version 1.8, you commited this:
        - Rename is_content() to has_content(), because that's what it tells
        us. Removed _is_content member bool - we can check _content.empty()
        instead.

Well, my anderstanding of that is different : the accessor had, for me, the 
right name. If we look at the tree generated by libxml, the correponding node 
has a particular type, XML_TEXT_NODE, and cannot have any attribute/children.
A xml code like "<name>foo</name>" will generate 2 nodes. One named "name", 
and another which IS the content ("foo").

Do you agree or did I get something wrong ?

Anyway I'm thinking of creating a special class for content nodes.

The thing I don't know yet is : can 'normal' node has both content node AND 
'normal' children ?

Regards,

Christophe
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iEYEARECAAYFAj3s0sMACgkQB+sU3TyOQjDtWwCg6y6kOnyW3nJ/9e80jEqDoHSD
QHoAn2r50FZgFFaiqX5SQgT+4Xzwu0Q9
=4GYm
-----END PGP SIGNATURE-----

[Libxmlplusplus-general] SaxParser and exceptions

[Christophe de V. <cde...@al...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi all,

It is now possible to throw exceptions from a callback method in the 
SaxParser.
The only condition is to throw an exception which herit from xmlpp::exception 
and to implement 2 virtual methods : Raise() and Clone().
An example is included in the CVS version.

If you have any comments about this do not hesitate...

Best regards,

Christophe


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (GNU/Linux)

iD8DBQE966k8B+sU3TyOQjARAhzmAJ40lP8ARYcuOkGbZTw8LMNrf5MQdwCgi94o
nSm86VhMNM+DHI0yQiqTrMA=
=oans
-----END PGP SIGNATURE-----

[Libxmlplusplus-general] Exceptions / Abort parsing

[<chr...@al...>]

Hi!

I tryed to throw an exception in a SaxParser::on_* function, but the
program dies (receiving SIGABRT) instead of catching it. Isn't this
possible because of libxml? And is there another way to stop parsing
within these callback functions?

regards,
Christian Neumann


-- 
GnuPG: pub 1024D/D8A91B4C 2001-11-11 Christian Neumann <chr...@al...>

Re: [Libxmlplusplus-general] Compilation errors with libxml++-0.16

[Christophe de V. <cde...@al...>]

=2D----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Le Mercredi 27 Novembre 2002 17:11, Mehdi Ouchia a =E9crit :
> Hi,

Hi,

>
> i try to use the latest version libxml++(0.16) with
> libxml2-2.4.0. Unfortunately, i've got several errors
> due to libxml++ compilation. Indeed, for example
> XML_GET_LINE() or XML_GET_CONTENT() aren't defined
> anywhere, neither in libxml2-2.4.0 or libxml++-0.16.

This is because these macros were introduced in a later version of libxml2.=
=20
Please try with the lastest release.

>
> Thanks in advance for your help.

You're welcome,

Christophe
=2D----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iEYEARECAAYFAj3k7/wACgkQB+sU3TyOQjAyfgCgspfGgbiJS5CZS4JkNFWCblFR
KjcAoK3P3ZplgWg3EU626GsqpeI7wS8q
=3DRc5J
=2D----END PGP SIGNATURE-----

[Libxmlplusplus-general] Compilation errors with libxml++-0.16

[Mehdi O. <meh...@ya...>]

Hi,
 
i try to use the latest version libxml++(0.16) with
libxml2-2.4.0. Unfortunately, i've got several errors
due to libxml++ compilation. Indeed, for example
XML_GET_LINE() or XML_GET_CONTENT() aren't defined
anywhere, neither in libxml2-2.4.0 or libxml++-0.16.

Thanks in advance for your help.

Best regards, 

 Mehdi.


__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com

Re: [Libxmlplusplus-general] DomParser and SaxParser

[Edd D. <ed...@us...>]

On Thu, 2002-11-21 at 17:33, Christophe de VIENNE wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>=20
> Hi,
>=20
> I think in a soon future we should reimplement the DomParser on top of th=
e Sax=20
> Parser.
> It may be a bit more difficult, but it would avoid to have this temporary=
=20
> libxml doc in memory, and especially all the nodes of the tree. We would=20
> construct directly our Nodes structure.
> On small documents the current implementation is not a problem, but with =
big=20
> ones it could save a LOT of memory...
>=20
> Any comments ?
>=20

Makes a lot of sense.=20

The current DOM implementation is a bit disappointing in that it doesn't
provide namespace information. I presume DOM Core level 2 would be a
good target to head for.

It would be most convenient to do it on top of a SAX2 implementation in
order to get the namespace information, rather than redoing the
namespace handling purely for the DOM layer.  Of course, a SAX2 API
would make me very happy.

-- Edd

Re: [Libxmlplusplus-general] DomParser and SaxParser

[Christophe de V. <cde...@al...>]

=2D----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Le Jeudi 21 Novembre 2002 18:33, Christophe de VIENNE a =E9crit :
> but with big
> ones it could save a LOT of memory...


And should be also much faster...
=2D----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iEYEARECAAYFAj3dGZ0ACgkQB+sU3TyOQjC7mACgsVzxlQNbfr/f7dkwlSAPrt3W
AFkAniaPmay0jd097Gh7lVv0qkw3NzZI
=3D8U0Z
=2D----END PGP SIGNATURE-----

[Libxmlplusplus-general] DomParser and SaxParser

[Christophe de V. <cde...@al...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,

I think in a soon future we should reimplement the DomParser on top of the Sax 
Parser.
It may be a bit more difficult, but it would avoid to have this temporary 
libxml doc in memory, and especially all the nodes of the tree. We would 
construct directly our Nodes structure.
On small documents the current implementation is not a problem, but with big 
ones it could save a LOT of memory...

Any comments ?


Christophe

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iEYEARECAAYFAj3dGP8ACgkQB+sU3TyOQjBFmgCfVp+lX+FzBlEoRwkkeZkO+RVw
yugAnjWmV3Gyj006HkvQi9uuAhfxGDXB
=Gs2S
-----END PGP SIGNATURE-----

[Libxmlplusplus-general] Parser::parse_stream(istream& in) added.

[Christophe de V. <cde...@al...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,

This method permit to parse from a stream and is implemented using the push 
method so it should consume low memory.
It has been tested with success on the examples.

Cheers,

Christophe
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iEYEARECAAYFAj3dF4YACgkQB+sU3TyOQjDcCgCgiuilhvhYhnZ8omuFGY15doSO
XG4AoKcYVVq5S8VQS9HAxKACwkOqKyRh
=5GhN
-----END PGP SIGNATURE-----