libxmlplusplus-general

[libxml++] [PATCH] clear_internal_subset()

[Thomas J. <tho...@in...>]

Hi,

I wanted to replace a dtd reference in a parsed document with a new one. Just 
calling set_internal_subset doesn't work, libxml checks if we already have at 
dtd and breaks.

So I wrote the clear_internal_subset function.

But since I haven't done much libxml hacking yet, I'm not sure if I got all 
the node linking, referencing and memory allocating right. I would highly 
appreciate it if a libxml guru could take a look at it.
Feel free to include this patch to libxml++.

Kind regards,
Thomas

Re: [libxml++] [PATCH] clear_internal_subset()

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

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

Hi,

Thanks for the patch.

A few little comments :
 - Why returning a bool ? I think only the post-condition of the function i=
s=20
important, which is always the same : no subset in the current document. If=
=20
one needs to know if a subset was present before, then he should check it=20
before. Otherwise if the absence of an internal subset is considered as an=
=20
error, then it should throw an exception (I don't think so).
 - I don't know if the implementation is correct (I seems to me that yes bu=
t=20
still). If nobody gives you a clear answer on this list, I'd suggest you to=
=20
ask on the libxml list.
 - I'd prefer you to use the patch manager next time. Thanks !


Regards,

Christophe

Le Mardi 9 Septembre 2003 10:13, Thomas Jarosch a =E9crit :
> Hi,
>
> I wanted to replace a dtd reference in a parsed document with a new one.
> Just calling set_internal_subset doesn't work, libxml checks if we already
> have at dtd and breaks.
>
> So I wrote the clear_internal_subset function.
>
> But since I haven't done much libxml hacking yet, I'm not sure if I got a=
ll
> the node linking, referencing and memory allocating right. I would highly
> appreciate it if a libxml guru could take a look at it.
> Feel free to include this patch to libxml++.
>
> Kind regards,
> Thomas
=2D----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)

iD8DBQE/XZUxB+sU3TyOQjARAkypAKCJaVeKKv0tVRmRrwdYRhdtgIrWzACgrJvc
iFZIUDo7zL1/1W4QGHeCq34=3D
=3DJveW
=2D----END PGP SIGNATURE-----

Re: [libxml++] [PATCH] clear_internal_subset()

[Thomas J. <tho...@in...>]

>  - Why returning a bool ? I think only the post-condition of the function
> is important, which is always the same : no subset in the current document.
> If one needs to know if a subset was present before, then he should check
> it before. Otherwise if the absence of an internal subset is considered as
> an error, then it should throw an exception (I don't think so).

Ok, it can be removed.

>  - I don't know if the implementation is correct (I seems to me that yes
> but still). If nobody gives you a clear answer on this list, I'd suggest
> you to ask on the libxml list.

Hmm, looks like I have to try my luck there.

>  - I'd prefer you to use the patch manager next time. Thanks !

Ok, sorry, next time.

One other thing: Is there an easy way to get rid of the error output
to stderr if a parser error occures (f.e. when I use binary garbage as input)?

Cheers,
Thomas

Re: [libxml++] [PATCH] clear_internal_subset()

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

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

Le Mardi 9 Septembre 2003 16:10, Thomas Jarosch a =E9crit :
> One other thing: Is there an easy way to get rid of the error output
> to stderr if a parser error occures (f.e. when I use binary garbage as
> input)?

I guess you're speaking about the output of libxml2 itself since as far as =
I=20
know there shouldn't be direct output done by libxml++. Well, this has to b=
e=20
investigated. I think this has been discussed on the libxml mailing-list=20
already. A search on xmlsolft site should give some hint.=20
If the solution is elegant (I mean is not a dirty hack), we could think abo=
ut=20
a wrapper in libxml++.

Cheers,

Christophe
=2D----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)

iD8DBQE/XfLrB+sU3TyOQjARApEGAJ48kO3gAnMfQMYOi7Tu45FRMzC/QACbBqKB
4sxVtfAUF4lAudZydortxns=3D
=3DEMeJ
=2D----END PGP SIGNATURE-----