Re: [Podofo-users] About fix for CVE-2017- 6845 (Was: Re: Next PoDoFo Release 0.9.6)

2018-02-22 Thread zyx
On Thu, 2018-02-22 at 09:38 +0100, Dominik Seichter wrote: > Would you be so kind and remove the #ifdef NDEBUG? i do not have > acess to svn right now. > Hi, sure thing, done as revision 1892: http://sourceforge.net/p/podofo/code/1892 Bye, zyx ---

Re: [Podofo-users] About fix for CVE-2017- 6845 (Was: Re: Next PoDoFo Release 0.9.6)

2018-02-22 Thread Dominik Seichter via Podofo-users
Hi zyx Yes that makes sense. I was also wandering why we do have logic checks only in debug. Would you be so kind and remove the #ifdef NDEBUG? i do not have acess to svn right now. Thanks Dominik Am 22.02.2018 09:30 schrieb "zyx" : > On Wed, 2018-02-21 at 21:51 +0100, Dominik Seichter wrote:

Re: [Podofo-users] About fix for CVE-2017- 6845 (Was: Re: Next PoDoFo Release 0.9.6)

2018-02-22 Thread zyx
On Wed, 2018-02-21 at 21:51 +0100, Dominik Seichter wrote: > It should always be legal to call this: > PODOFO_RAISE_LOGIC_IF( m_stack.empty(), "Can get current > graphicsstate!" ); > > As it translates to: > if( m_stack.empty() ) { > throw ... > } Hi, the macro is fine, the problem is wh

Re: [Podofo-users] About fix for CVE-2017- 6845 (Was: Re: Next PoDoFo Release 0.9.6)

2018-02-21 Thread Dominik Seichter via Podofo-users
Hi zyx, Thanks for the solid analysis. I agree and reverted the fix I did in 1873. I am not sure how we PODOFO_RAISE_LOGIC_IF can cause such an error? It should always be legal to call this: PODOFO_RAISE_LOGIC_IF( m_stack.empty(), "Can get current graphicsstate!" ); As it translates to: if( m_st