Never mind ... problem solved. In the course of putting a complete
example, I discovered that my Makefile had
-I/usr/local/openbabel/lib/openbabel-2.0
which points to the OpenBabel 2.2.x code. In fact I was linking against a
new version of OpenBabel 2.3.x, so the include files didn't match the
libraries.
Craig
On Mon, May 7, 2012 at 11:05 AM, Craig James <cja...@emolecules.com> wrote:
> On Mon, May 7, 2012 at 10:09 AM, Geoffrey Hutchison <geo...@pitt.edu>wrote:
>
>>
>> On May 4, 2012, at 10:45 PM, Craig James wrote:
>>
>> > But it seems to be the atom iterator, not the ring iterator, that's
>> causing the problem.
>>
>> Huh. Here's comparable code in forcefieldmmff94.cpp:
>>
>> for(rj = (*ri)->_path.begin();rj != (*ri)->_path.end();rj++) { //
>> for each ring atom
>> index = *rj;
>> ringatom = _mol.GetAtom(index);
>>
>> Granted, like Igor's comment the code gets the rings like so:
>>
>> vector<OBRing*> vr;
>> vr = _mol.GetSSSR();
>>
>> Have you checked the path size?
>>
>
> It's a random number. Here is a more concise version of the code ... it's
> extracted from a much larger body of code, but I think I got all the
> relevant parts.
>
>
> OBMol *pmol;
> OBConversion conv;
> conv.SetInAndOutFormats("sdf", "smi");
> pmol = new OBMol;
> pmol->Clear();
> conv.ReadString(pmol, molData_str);
>
> natoms = pmol->NumAtoms();
> nbonds = pmol->NumBonds();
>
> if (natoms <= 0) { // No atoms? (shouldn't happen...)
> return 0;
> }
>
> for (atom = pmol->BeginAtom(ai); atom; atom = pmol->NextAtom(ai)) {
> float x = atom->GetX();
> float y = atom->GetY();
> // --- snip --- code to draw atom labels into image
> }
>
>
> // --- snip --- initialize code for GIF/JPG/PNG library
>
> vector<OBRing*> rlist = mol->GetSSSR();
> vector<OBRing*>::iterator ri;
>
> for (ri = rlist.begin(); ri != rlist.end(); ++ri) {
>
> fprintf(stderr, "_path.size() = %d\n", (*ri)->_path.size());
> // Prints out a big random negative number
>
> vector<int>::iterator j;
> for (j = (*ri)->_path.begin(); j != (*ri)->_path.end(); j++) {
> OBAtom *atom = mol->GetAtom(*j);
> // .... crashes because j is NULL
> }
> }
>
> I ran valgrind on it which is usually very good at finding problems. When
> it gets to where I print _path.size(), valgrind complains that I'm printing
> uninitialized memory, and the value that comes out is indeed a random value
> (-1067674809 or 0x3FA36CB9). valgrind doesn't report any other errors at
> all.
>
> I'm stumped, but I'm happy to help,
>>
>
> Thanks ... I suspect this will turn out to be a bug somewhere else in the
> program (overwriting a buffer or something, but in a way that valgrind
> can't detect), but for the life of me I can't find it. It worked for
> OpenBabel 2.2.x and now fails for OpenBabel 2.3.x.
>
> Mostly I'm just asking if I'm calling the OpenBabel functions correctly.
> It sounds like I am. That being the case, I'll start digging around in the
> rest of my code to find the bug.
>
> Craig
>
>
------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobile security and the latest in malware
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
OpenBabel-Devel mailing list
OpenBabel-Devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/openbabel-devel
