You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@santuario.apache.org by Vitaly Prapirny <ma...@mebius.net> on 2007/08/22 11:52:08 UTC

XML Security C++: Incomplete cleanup in XSECC14n20010315 destructor

Hi!

XSECC14n20010315 destructor performs delete [] m_exclNSList[i] but
m_exclNSList items was allocated with strdup and must be released
with free(). And mp_attributes cleanup is absent in code. So I
propose this version of destructor:

XSECC14n20010315::~XSECC14n20010315() {

	if (mp_formatter != NULL)
		delete mp_formatter;

	// Clear out the exclusive namespace list
	int size = (int) m_exclNSList.size();

	for (int i = 0; i < size; ++i) {

		free(m_exclNSList[i]);

	}

	m_exclNSList.clear();

	while (mp_attributes != NULL) {

		mp_currentAttribute = mp_attributes->next;
		delete mp_attributes;
		mp_attributes = mp_currentAttribute;
	}

	mp_attributes = mp_currentAttribute = mp_firstNonNsAttribute = NULL;
}

Good luck!
	Vitaly

Re: XML Security C++: Incomplete cleanup in XSECC14n20010315 destructor

Posted by Vitaly Prapirny <ma...@mebius.net>.
Berin Lautenbach wrote:
> Hmm.  That's an interesting one.  Have you seen a case where the 
> mp_attributes list is not already clear when the destructor is called?

Yes. That is why this fix has been appeared. But I can't remember those
particular case for now, after almost three years :) Sorry for late
report.

> I can see how it could occur if the canonicalisation gets interrupted in 
> the midst of outputting Attribute nodes.  This fix would work nicely, 
> and I think it should be safe as the code that normally cleans up 
> ensures mp_attributes is NULL, so it won't get called unnecessarily.

I reached the same conclusion when focusing on this issue today.

Good luck!
	Vitaly


RE: XML Security C++: Incomplete cleanup in XSECC14n20010315 destructor

Posted by Scott Cantor <ca...@osu.edu>.
> I can see how it could occur if the canonicalisation gets interrupted in
> the midst of outputting Attribute nodes.  This fix would work nicely,
> and I think it should be safe as the code that normally cleans up
> ensures mp_attributes is NULL, so it won't get called unnecessarily.

Ok, I checked this in also.

I will scan the bugzilla, as well, but anybody willing to check out the
trunk and build it on their favorite OS, particular with Xalan and/or the
NSS support would be appreciated.

-- Scott



Re: XML Security C++: Incomplete cleanup in XSECC14n20010315 destructor

Posted by Berin Lautenbach <be...@wingsofhermes.org>.
Hmm.  That's an interesting one.  Have you seen a case where the 
mp_attributes list is not already clear when the destructor is called?

I can see how it could occur if the canonicalisation gets interrupted in 
the midst of outputting Attribute nodes.  This fix would work nicely, 
and I think it should be safe as the code that normally cleans up 
ensures mp_attributes is NULL, so it won't get called unnecessarily.

Cheers,
	Berin

Vitaly Prapirny wrote:
> Hi!
> 
> XSECC14n20010315 destructor performs delete [] m_exclNSList[i] but
> m_exclNSList items was allocated with strdup and must be released
> with free(). And mp_attributes cleanup is absent in code. So I
> propose this version of destructor:
> 
> XSECC14n20010315::~XSECC14n20010315() {
> 
>     if (mp_formatter != NULL)
>         delete mp_formatter;
> 
>     // Clear out the exclusive namespace list
>     int size = (int) m_exclNSList.size();
> 
>     for (int i = 0; i < size; ++i) {
> 
>         free(m_exclNSList[i]);
> 
>     }
> 
>     m_exclNSList.clear();
> 
>     while (mp_attributes != NULL) {
> 
>         mp_currentAttribute = mp_attributes->next;
>         delete mp_attributes;
>         mp_attributes = mp_currentAttribute;
>     }
> 
>     mp_attributes = mp_currentAttribute = mp_firstNonNsAttribute = NULL;
> }
> 
> Good luck!
>     Vitaly
> 
> 

RE: XML Security C++: Incomplete cleanup in XSECC14n20010315 destructor

Posted by Scott Cantor <ca...@osu.edu>.
> XSECC14n20010315 destructor performs delete [] m_exclNSList[i] but
> m_exclNSList items was allocated with strdup and must be released
> with free(). And mp_attributes cleanup is absent in code. So I
> propose this version of destructor:

I'll check in the free fix, but I'd like to get confirmation from Berin that
the rest is accurate, I don't even begin to follow the c14n code and I've
never noticed any significant leaks from verifying signatures, so it would
be nice to get a sanity check that something else isn't freeing that
information somehow.

-- Scott