int ReadPQS_geom(istream &ifs, OBMol &mol, const char *title,
int input_style, double bohr_to_angstrom)
{
int atom_count=0;
double x, y, z;
char buffer[BUFF_SIZE];
string str;
OBAtom *atom;
vector<string> vs;
mol.Clear();
mol.BeginModify();
while (ifs.getline(buffer,BUFF_SIZE) && !card_found(buffer))
{
if (buffer[0]!='$')
{
tokenize(vs, buffer);
if (vs.size() < 1) return false; // timvdm 18/06/2008
atom=mol.NewAtom();
str=vs[0];
if (input_style==0)
{
if (vs.size() < 4) return false; // timvdm 18/06/2008
atom->SetAtomicNum(OBElements::GetAtomicNum(str.c_str()));
x=atof((char*) vs[1].c_str())*bohr_to_angstrom;
y=atof((char*) vs[2].c_str())*bohr_to_angstrom;
z=atof((char*) vs[3].c_str())*bohr_to_angstrom;
}
else
{
if (vs.size() < 5) return false; // timvdm 18/06/2008
str.replace (0,2,"");
atom->SetAtomicNum(OBElements::GetAtomicNum(str.c_str()));
x=atof((char*) vs[2].c_str())*bohr_to_angstrom;
y=atof((char*) vs[3].c_str())*bohr_to_angstrom;
z=atof((char*) vs[4].c_str())*bohr_to_angstrom;
}
atom->SetVector(x, y, z);
atom_count++;
}
}
mol.ConnectTheDots();
mol.PerceiveBondOrders();
mol.EndModify();
mol.SetTitle(title);
return atom_count;
}
/* FUNCTION: basicExportTest */
void OpenBabelImportExportTest::basicExportTest() {
// Create a water molecule for frame 0
//
int frameSetId = entityManager->addFrameSet();
int frameIndex = entityManager->addFrame(frameSetId);
NXMoleculeSet* rootMoleculeSet =
entityManager->getRootMoleculeSet(frameSetId, frameIndex);
OBMol* molecule = rootMoleculeSet->newMolecule();
OBAtom* atomO = molecule->NewAtom();
atomO->SetAtomicNum(etab.GetAtomicNum("O")); // Oxygen
atomO->SetVector(0.00000000, 0.00000000, 0.37000000); // Angstroms
OBAtom* atomH1 = molecule->NewAtom();
atomH1->SetAtomicNum(etab.GetAtomicNum("H")); // Hydrogen
atomH1->SetVector(0.78000000, 0.00000000, -0.18000000);
OBAtom* atomH2 = molecule->NewAtom();
atomH2->SetAtomicNum(etab.GetAtomicNum("H")); // Hydrogen
atomH2->SetVector(-0.78000000, 0.00000000, -0.18000000);
OBBond* bond = molecule->NewBond();
bond->SetBegin(atomO);
bond->SetEnd(atomH1);
bond = molecule->NewBond();
bond->SetBegin(atomO);
bond->SetEnd(atomH2);
// Write it with the OpenBabelImportExport plugin
NXCommandResult* commandResult =
entityManager->exportToFile("testOpenBabel.pdb", frameSetId, -1);
if (commandResult->getResult() != NX_CMD_SUCCESS)
printf("\n%s\n", qPrintable(GetNV1ResultCodeString(commandResult)));
CPPUNIT_ASSERT(commandResult->getResult() == NX_CMD_SUCCESS);
}
bool ReadFeat(istream &ifs,OBMol &mol, const char *title)
{
char buffer[BUFF_SIZE];
int i,natoms;
ifs.getline(buffer,BUFF_SIZE);
sscanf(buffer,"%d",&natoms);
mol.ReserveAtoms(natoms);
mol.BeginModify();
if (!ifs.getline(buffer,BUFF_SIZE)) return(false);
mol.SetTitle(buffer);
double x,y,z;
char type[20];
OBAtom *atom;
for (i = 0; i < natoms;i++)
{
if (!ifs.getline(buffer,BUFF_SIZE)) return(false);
sscanf(buffer,"%s %lf %lf %lf",
type,
&x,
&y,
&z);
CleanAtomType(type);
atom = mol.NewAtom();
atom->SetVector(x,y,z);
atom->SetAtomicNum(etab.GetAtomicNum(type));
}
mol.EndModify();
return(true);
}
bool PubChemFormat::EndElement(const string& name)
{
unsigned int i;
if(name=="PC-Atoms")
{
for(i=0;i<AtNum.size();++i)
{
OBAtom* pAtom = _pmol->NewAtom();
pAtom->SetAtomicNum(AtNum[i]);
}
}
else if(name=="PC-Bonds")
{
for(i=0;i<BondBeginAtIndx.size();++i)
_pmol->AddBond(BondBeginAtIndx[i],BondEndAtIndx[i],BondOrder[i]);
}
else if(name=="PC-Conformer")
{
++ConformerIndx;
if(Coordz.size()!=Coordx.size())
Coordz.resize(Coordx.size());
for(i=0;i<CoordIndx.size();++i)
{
OBAtom* pAtom = _pmol->GetAtom(CoordIndx[i]);
pAtom->SetVector(Coordx[i],Coordy[i],Coordz[i]);
}
}
else if(name=="PC-Compound") //this is the end of the molecule we are extracting
{
_pmol->EndModify();
return false;//means stop parsing
}
return true;
}
/**
Method reads coordinates from input stream (ifs) and
writes it into supplied OBMol object (molecule).
Input stream must be set to begining of coordinates
table in nwo file. (Line after "Output coordinates...")
Stream will be set at next line after geometry table.
If one of input arguments is NULL method returns without
any changes
If "molecule" already contain geometry - method will write
new geometry as conformer.
If "molecule" contain geometry which incompatible with read
data method returns without changes.
*/
void NWChemOutputFormat::ReadCoordinates(istream* ifs, OBMol* molecule)
{
if ((molecule == NULL) || (ifs == NULL))
return;
vector<string> vs;
char buffer[BUFF_SIZE];
double x, y, z;
unsigned int natoms = molecule->NumAtoms();
bool from_scratch = false;
double* coordinates;
if (natoms == 0)
from_scratch = true;
else
coordinates = new double[3*natoms];
ifs->getline(buffer,BUFF_SIZE); // blank
ifs->getline(buffer,BUFF_SIZE); // column headings
ifs->getline(buffer,BUFF_SIZE); // ---- ----- ----
ifs->getline(buffer,BUFF_SIZE);
tokenize(vs,buffer);
unsigned int i=0;
while (vs.size() == 6)
{
x = atof((char*)vs[3].c_str());
y = atof((char*)vs[4].c_str());
z = atof((char*)vs[5].c_str());
if (from_scratch)
{
// set atomic number
OBAtom* atom = molecule->NewAtom();
atom->SetAtomicNum(atoi(vs[2].c_str()));
atom->SetVector(x,y,z);
}
else
{
// check atomic number
if ((i>=natoms) || (molecule->GetAtom(i+1)->GetAtomicNum() != atoi(vs[2].c_str())))
{
delete[] coordinates;
return;
}
coordinates[i*3] = x;
coordinates[i*3+1] = y;
coordinates[i*3+2] = z;
i++;
}
if (!ifs->getline(buffer,BUFF_SIZE))
break;
tokenize(vs,buffer);
}
if (from_scratch)
{
return;
}
if (i != natoms) {
delete[] coordinates;
return;
}
molecule->AddConformer(coordinates);
}
/**
Method reads partial charges from input stream (ifs)
and writes them to supplied OBMol object (molecule)
Input stream must be set to begining of charges
table in nwo file. (Line after "Mulliken analysis of the total density")
Stream will be set at next line after charges table.
If reading charges failed or "molecule" contains
data incompatible with read charges then "molecule"
wont be changed.
*/
void NWChemOutputFormat::ReadPartialCharges(istream* ifs, OBMol* molecule)
{
if ((molecule == NULL) || (ifs == NULL))
return;
vector<string> vs;
char buffer[BUFF_SIZE];
bool from_scratch = false;
vector<int> charges;
vector<double> partial_charges;
unsigned int natoms = molecule->NumAtoms();
if (natoms == 0)
from_scratch = true;
ifs->getline(buffer,BUFF_SIZE); // ---- ----- ----
ifs->getline(buffer,BUFF_SIZE); // blank
ifs->getline(buffer,BUFF_SIZE); // column headings
ifs->getline(buffer,BUFF_SIZE); // ---- ----- ----
ifs->getline(buffer,BUFF_SIZE);
tokenize(vs, buffer);
// N Symbol Charge PartialCharge+Charge ShellCharges
// 0 1 2 3 4,etc
unsigned int i = 1;
while (vs.size() >= 4)
{
int charge = atoi(vs[2].c_str());
if (!from_scratch)
{
if (i > natoms)
return;
if (molecule->GetAtom(i++)->GetAtomicNum() != charge)
return;
}
else
charges.push_back(charge);
partial_charges.push_back(atof(vs[3].c_str()) - charge);
ifs->getline(buffer,BUFF_SIZE);
tokenize(vs, buffer);
}
if (from_scratch)
molecule->ReserveAtoms(partial_charges.size());
else if (partial_charges.size() != natoms)
return;
for(unsigned int j=0;j<partial_charges.size();j++)
{
OBAtom* atom;
if (from_scratch)
{
atom = molecule->NewAtom();
atom->SetAtomicNum(charges[j]);
}
else
{
atom = molecule->GetAtom(j+1);
}
atom->SetPartialCharge(partial_charges[j]);
}
}
bool AmberPrepFormat::ReadMolecule(OBBase* pOb, OBConversion* pConv)
{
OBMol* pmol = pOb->CastAndClear<OBMol>();
if(pmol==NULL)
return false;
//Define some references so we can use the old parameter names
istream &ifs = *pConv->GetInStream();
OBMol &mol = *pmol;
const char* title = pConv->GetTitle();
char buffer[BUFF_SIZE];
string str,str1;
OBAtom *atom;
OBInternalCoord *coord;
vector<string> vs;
vector<OBInternalCoord*> internals;
mol.BeginModify();
while (ifs.getline(buffer,BUFF_SIZE))
{
tokenize(vs,buffer);
if (vs.size() == 10)
{
atom = mol.NewAtom();
coord = new OBInternalCoord();
if (mol.NumAtoms() > 1)
coord->_a = mol.GetAtom(atoi(vs[4].c_str()));
if (mol.NumAtoms() > 2)
coord->_b = mol.GetAtom(atoi(vs[5].c_str()));
if (mol.NumAtoms() > 3)
coord->_c = mol.GetAtom(atoi(vs[6].c_str()));
coord->_dst = atof(vs[7].c_str());
coord->_ang = atof(vs[8].c_str());
coord->_tor = atof(vs[9].c_str());
internals.push_back(coord);
atom->SetAtomicNum(etab.GetAtomicNum(vs[1].c_str()));
if (!ifs.getline(buffer,BUFF_SIZE))
break;
tokenize(vs,buffer);
}
}
if (internals.size() > 0)
InternalToCartesian(internals,mol);
if (!pConv->IsOption("b",OBConversion::INOPTIONS))
mol.ConnectTheDots();
if (!pConv->IsOption("s",OBConversion::INOPTIONS) && !pConv->IsOption("b",OBConversion::INOPTIONS))
mol.PerceiveBondOrders();
mol.EndModify();
mol.SetTitle(title);
return(true);
}
bool BallStickFormat::ReadMolecule(OBBase* pOb, OBConversion* pConv)
{
OBMol* pmol = pOb->CastAndClear<OBMol>();
if(pmol==NULL)
return false;
//Define some references so we can use the old parameter names
istream &ifs = *pConv->GetInStream();
OBMol &mol = *pmol;
const char* title = pConv->GetTitle();
int i,natoms;
char buffer[BUFF_SIZE];
if (!ifs.getline(buffer,BUFF_SIZE))
return(false);
if (!ifs.getline(buffer,BUFF_SIZE))
return(false);
sscanf(buffer,"%d",&natoms);
mol.ReserveAtoms(natoms);
mol.BeginModify();
double x,y,z;
OBAtom *atom;
vector<string> vs;
vector<string>::iterator j;
for (i = 1; i <= natoms;i ++)
{
if (!ifs.getline(buffer,BUFF_SIZE))
return(false);
tokenize(vs,buffer);
if (vs.size() < 4)
return(false);
if (vs[0].size() > 1)
vs[0][1] = tolower(vs[0][1]);
atom = mol.NewAtom();
x = atof((char*)vs[1].c_str());
y = atof((char*)vs[2].c_str());
z = atof((char*)vs[3].c_str());
atom->SetVector(x,y,z); //set coordinates
atom->SetAtomicNum(etab.GetAtomicNum(vs[0].c_str()));
for (j = vs.begin()+4;j != vs.end();j++)
mol.AddBond(atom->GetIdx(),atoi((char*)j->c_str()),1);
}
// clean out any remaining blank lines
while(ifs.peek() != EOF && ifs.good() &&
(ifs.peek() == '\n' || ifs.peek() == '\r'))
ifs.getline(buffer,BUFF_SIZE);
mol.EndModify();
mol.SetTitle(title);
return(true);
}
bool ReadAlchemy(istream &ifs,OBMol &mol, const char *title)
{
int i;
int natoms,nbonds;
char buffer[BUFF_SIZE];
ifs.getline(buffer,BUFF_SIZE);
sscanf(buffer,"%d %*s %d", &natoms, &nbonds);
if (!natoms) return(false);
mol.ReserveAtoms(natoms);
ttab.SetFromType("ALC");
string str;
double x,y,z;
OBAtom *atom;
vector<string> vs;
for (i = 1; i <= natoms; i ++)
{
if (!ifs.getline(buffer,BUFF_SIZE)) return(false);
tokenize(vs,buffer);
if (vs.size() != 6) return(false);
atom = mol.NewAtom();
x = atof((char*)vs[2].c_str());
y = atof((char*)vs[3].c_str());
z = atof((char*)vs[4].c_str());
atom->SetVector(x,y,z); //set coordinates
//set atomic number
ttab.SetToType("ATN"); ttab.Translate(str,vs[1]);
atom->SetAtomicNum(atoi(str.c_str()));
//set type
ttab.SetToType("INT"); ttab.Translate(str,vs[1]);
atom->SetType(str);
}
char bobuf[100];
string bostr;
int bgn,end,order;
for (i = 0; i < nbonds; i++)
{
if (!ifs.getline(buffer,BUFF_SIZE)) return(false);
sscanf(buffer,"%*d%d%d%s",&bgn,&end,bobuf);
bostr = bobuf; order = 1;
if (bostr == "DOUBLE") order = 2;
else if (bostr == "TRIPLE") order = 3;
else if (bostr == "AROMATIC") order = 5;
mol.AddBond(bgn,end,order);
}
mol.SetTitle(title);
return(true);
}
//------------------------------------------------------------------------------
bool OBMoldenFormat::ReadMolecule( OBBase* pOb, OBConversion* pConv )
{
OBMol* pmol = dynamic_cast< OBMol* >(pOb);
if( pmol == 0 ) return false;
istream& ifs = *pConv->GetInStream();
pmol->BeginModify();
pmol->SetDimension( 3 );
string lineBuffer;
getline( ifs, lineBuffer );
while( ifs && lineBuffer.find( "[Atoms]" ) == string::npos
&& lineBuffer.find( "[ATOMS]" ) == string::npos )
{
getline( ifs, lineBuffer );
}
if( !ifs ) return false;
//[Atoms] AU OR Angs
double factor = 1.; // Angstrom
if( lineBuffer.find( "AU" ) != string::npos ) factor = 0.529177249; // Bohr
while( ifs )
{
getline( ifs, lineBuffer );
if( lineBuffer.size() == 0 ) continue;
if( lineBuffer.find( '[' ) != string::npos ) break;
istringstream is( lineBuffer );
string atomName;
int atomId;
int atomicNumber;
double x, y, z;
is >> atomName >> atomId >> atomicNumber >> x >> y >> z;
OBAtom* atom = pmol->NewAtom();
if( !atom ) break;
atom->SetAtomicNum( atomicNumber );
atom->SetVector( x * factor, y * factor, z * factor );
}
if( !pConv->IsOption( "b", OBConversion::INOPTIONS ) ) pmol->ConnectTheDots();
if (!pConv->IsOption( "s", OBConversion::INOPTIONS )
&& !pConv->IsOption( "b", OBConversion::INOPTIONS ) )
{
pmol->PerceiveBondOrders();
}
pmol->EndModify();
return true;
}
bool ReadBiosymCAR(istream &ifs,OBMol &mol, const char *title)
{
char buffer[BUFF_SIZE];
string str;
double x,y,z;
OBAtom *atom;
vector<string> vs;
mol.BeginModify();
while (ifs.getline(buffer,BUFF_SIZE))
{
if(strstr(buffer,"PBC") != NULL)
{
if(strstr(buffer,"ON") != NULL)
{
ifs.getline(buffer,BUFF_SIZE);
ifs.getline(buffer,BUFF_SIZE);
ifs.getline(buffer,BUFF_SIZE);
}
else
{
ifs.getline(buffer,BUFF_SIZE);
ifs.getline(buffer,BUFF_SIZE);
}
break;
}
}
while (ifs.getline(buffer,BUFF_SIZE))
{
if(strstr(buffer,"end") != NULL)
break;
atom = mol.NewAtom();
tokenize(vs,buffer);
atom->SetAtomicNum(etab.GetAtomicNum(vs[7].c_str()));
x = atof((char*)vs[1].c_str());
y = atof((char*)vs[2].c_str());
z = atof((char*)vs[3].c_str());
atom->SetVector(x,y,z);
}
mol.EndModify();
mol.ConnectTheDots();
mol.PerceiveBondOrders();
mol.SetTitle(title);
return(true);
}
bool ReadHIN(istream &ifs,OBMol &mol, const char *title)
{
// Right now only read in the first molecule
int i;
int max, bo;
char buffer[BUFF_SIZE];
string str,str1;
double x,y,z;
OBAtom *atom;
vector<string> vs;
ifs.getline(buffer, BUFF_SIZE);
while (strstr(buffer,"mol") == NULL)
ifs.getline(buffer, BUFF_SIZE);
ifs.getline(buffer, BUFF_SIZE);
mol.BeginModify();
while (strstr(buffer,"endmol") == NULL)
{
tokenize(vs,buffer); // Don't really know how long it'll be
if (vs.size() < 11) break;
atom = mol.NewAtom();
atom->SetAtomicNum(etab.GetAtomicNum(vs[3].c_str()));
x = atof((char*)vs[7].c_str());
y = atof((char*)vs[8].c_str());
z = atof((char*)vs[9].c_str());
atom->SetVector(x,y,z);
max = 11 + 2 * atoi((char *)vs[10].c_str());
for (i = 11; i < max; i+=2)
{
switch(((char*)vs[i+1].c_str())[0]) // First char in next token
{
case 's': bo = 1; break;
case 'd': bo = 2; break;
case 't': bo = 3; break;
case 'a': bo = 5; break;
default : bo = 1; break;
}
mol.AddBond(mol.NumAtoms(), atoi((char *)vs[i].c_str()), bo);
}
ifs.getline(buffer, BUFF_SIZE);
}
mol.EndModify();
mol.SetTitle(title);
return(true);
}
void testIdsAddAtom()
{
OBMol mol;
// add 5 atoms
for (int i = 0; i < 5; ++i)
mol.NewAtom();
OBAtom a;
a.SetAtomicNum(6);
// add a sixth atom
mol.AddAtom(a);
OB_REQUIRE( mol.NumAtoms() == 6 );
OB_REQUIRE( mol.GetAtomById(5) );
OB_REQUIRE( mol.GetAtomById(5)->GetId() == 5 );
}
void AliasData::DeleteExpandedAtoms(OBMol& mol)
{
//The atom that carries the AliasData object remains as an Xx atom with no charge;
//the others are deleted. All the attached hydrogens are also deleted.
for(unsigned i=0;i<_expandedatoms.size();++i)
{
OBAtom* at = mol.GetAtomById(_expandedatoms[i]);
if(!at)
continue;
mol.DeleteHydrogens(at);
if(at->HasData(AliasDataType))
{
at->SetAtomicNum(0);
at->SetFormalCharge(0);
at->SetSpinMultiplicity(0);
}
else
mol.DeleteAtom(at);
}
_expandedatoms.clear();
}
bool MacroModFormat::ReadMolecule(OBBase* pOb, OBConversion* pConv)
{
OBMol* pmol = pOb->CastAndClear<OBMol>();
if(pmol==NULL)
return false;
//Define some references so we can use the old parameter names
istream &ifs = *pConv->GetInStream();
OBMol &mol = *pmol;
const char* defaultTitle = pConv->GetTitle();
// Get Title
char buffer[BUFF_SIZE];
int natoms;
vector<vector<pair<int,int> > > connections;
if (ifs.getline(buffer,BUFF_SIZE))
{
vector<string> vs;
tokenize(vs,buffer," \n");
if ( !vs.empty() && vs.size() > 0)
sscanf(buffer,"%i%*s",&natoms);
if (natoms == 0)
return false;
if ( !vs.empty() && vs.size() > 1)
mol.SetTitle(vs[1]);
else
{
string s = defaultTitle;
mol.SetTitle(defaultTitle);
}
}
else
return(false);
mol.BeginModify();
mol.ReserveAtoms(natoms);
connections.resize(natoms+1);
/***********************************************************************/
// Get Type Bonds, BondOrder, X, Y, Z
double x,y,z;
vector3 v;
char temp_type[10];
int i,j;
double charge;
OBAtom atom;
ttab.SetFromType("MMD");
for (i = 1; i <= natoms; i++)
{
if (!ifs.getline(buffer,BUFF_SIZE))
break;
int end[6], order[6];
sscanf(buffer,"%9s%d%d%d%d%d%d%d%d%d%d%d%d%lf%lf%lf",
temp_type,&end[0],&order[0],&end[1],&order[1],&end[2],&order[2],
&end[3], &order[3], &end[4], &order[4], &end[5], &order[5],
&x, &y, &z);
pair<int,int> tmp;
for ( j = 0 ; j <=5 ; j++ )
{
if ( end[j] > 0 && end[j] > i)
{
tmp.first = end[j];
tmp.second = order[j];
connections[i].push_back(tmp);
}
}
v.SetX(x);
v.SetY(y);
v.SetZ(z);
atom.SetVector(v);
string str = temp_type,str1;
ttab.SetToType("ATN");
ttab.Translate(str1,str);
atom.SetAtomicNum(atoi(str1.c_str()));
ttab.SetToType("INT");
ttab.Translate(str1,str);
atom.SetType(str1);
// stuff for optional fields
buffer[109]='\0';
sscanf(&buffer[101],"%lf", &charge);
atom.SetPartialCharge(charge);
mol.AddAtom(atom);
}
for (i = 1; i <= natoms; i++)
for (j = 0; j < (signed)connections[i].size(); j++)
mol.AddBond(i, connections[i][j].first, connections[i][j].second);
mol.EndModify();
OBBond *bond;
vector<OBBond*>::iterator bi;
for (bond = mol.BeginBond(bi);bond;bond = mol.NextBond(bi))
if (bond->GetBO() == 5 && !bond->IsInRing())
bond->SetBO(1);
if ( natoms != (signed)mol.NumAtoms() )
return(false);
// clean out remaining blank lines
while(ifs.peek() != EOF && ifs.good() &&
(ifs.peek() == '\n' || ifs.peek() == '\r'))
ifs.getline(buffer,BUFF_SIZE);
return(true);
}
bool ABINITFormat::ReadMolecule(OBBase* pOb, OBConversion* pConv)
{
OBMol* pmol = pOb->CastAndClear<OBMol>();
if(pmol==NULL)
return false;
//Define some references so we can use the old parameter names
istream &ifs = *pConv->GetInStream();
OBMol &mol = *pmol;
const char* title = pConv->GetTitle();
char buffer[BUFF_SIZE];
string str;
vector<string> vs;
OBAtom *atom;
int natom=-1;
vector<int> atomicNumbers, atomTypes;
double x, y, z;
vector<vector3> atomPositions;
vector<double> energies;
// Translation vectors (if present)
// aka rprim
vector3 translationVectors[3];
double acell[3]; // scale of lattice vectors
int numTranslationVectors = 0;
int symmetryCode = 0;
mol.BeginModify();
while (ifs.getline(buffer,BUFF_SIZE))
{
// tokens are listed in alphabetical order for code clarity
if (strstr(buffer, "acell")) {
tokenize(vs, buffer);
if (vs.size() < 4)
continue; // invalid line
// acell= 7.6967369631E+00 7.6967369631E+00 7.6967369631E+00
for (int i = 0; i < 3; ++i) {
acell[i] = atof(vs[i+1].c_str());
}
}
// Sometimes Cartesian has lower-case letter
else if
((strstr(buffer, "artesian coordinates")) &&
(strstr(buffer, "(xcart)")) &&
(natom > -1) //be sure natom has already been read
)
{
double unit = BOHR_TO_ANGSTROM;
if (strstr(buffer, "ngstrom"))
unit = 1.0; // no conversion needed
for (int i=0;i<natom;i++)
{
ifs.getline(buffer,BUFF_SIZE);
tokenize(vs, buffer);
if (vs.size() != 3) {
obErrorLog.ThrowError
(__FUNCTION__,
"error while parsing coordinates of atoms", obWarning);
return(false);
}
x = atof(vs[0].c_str()) * unit;
y = atof(vs[1].c_str()) * unit;
z = atof(vs[2].c_str()) * unit;
atomPositions.push_back(vector3(x, y, z));
// get next line
}
}
else if (strstr(buffer, "natom")) {
tokenize(vs, buffer);
if (vs.size() != 2)
continue;
natom = atoi(vs[1].c_str());
}
else if (strstr(buffer, "rprim")) {
numTranslationVectors = 0;
int column;
for (int i = 0; i < 3; ++i) {
tokenize(vs, buffer);
if (vs.size() < 3)
break;
// first line, rprim takes up a token
if (i == 0)
column = 1;
else
column = 0;
x = atof((char*)vs[column].c_str()) * BOHR_TO_ANGSTROM;
y = atof((char*)vs[column+1].c_str()) * BOHR_TO_ANGSTROM;
z = atof((char*)vs[column+2].c_str()) * BOHR_TO_ANGSTROM;
translationVectors[numTranslationVectors++].Set(x, y,z);
ifs.getline(buffer,BUFF_SIZE);
}
}
else if (strstr(buffer, "Symmetries")) {
tokenize(vs, buffer, "()");
// Should be something like (#160)
symmetryCode = atoi(vs[1].substr(1).c_str());
}
else if (strstr(buffer, "typat")) {
tokenize(vs, buffer);
atomTypes.clear();
for (unsigned int i = 1; i < vs.size(); ++i) {
atomTypes.push_back(atoi(vs[i].c_str()));
}
}
else if (strstr(buffer, "znucl")) {
tokenize(vs, buffer);
// make sure znucl is first token
if (vs[0] != "znucl")
continue;
// push back the remaining tokens into atomicNumbers
atomicNumbers.clear();
atomicNumbers.push_back(0); // abinit starts typat with 1
for (unsigned int i = 1; i < vs.size(); ++i)
atomicNumbers.push_back(int(atof(vs[i].c_str())));
}
// xangst
// forces
}
for (int i = 0; i < natom; ++i) {
atom = mol.NewAtom();
//set atomic number
int idx = atom->GetIdx();
int type = atomTypes[idx - 1];
atom->SetAtomicNum(atomicNumbers[type]);
// we set the coordinates by conformers in another loop
}
mol.EndModify();
int numConformers = atomPositions.size() / natom;
for (int i = 0; i < numConformers; ++i) {
double *coordinates = new double[natom * 3];
for (int j = 0; j < natom; ++j) {
vector3 currentPosition = atomPositions[i*natom + j];
coordinates[j*3] = currentPosition.x();
coordinates[j*3 + 1] = currentPosition.y();
coordinates[j*3 + 2] = currentPosition.z();
}
mol.AddConformer(coordinates);
}
// Delete first conformer, created by EndModify, bunch of 0s
mol.DeleteConformer(0);
// Set geometry to last one
mol.SetConformer(mol.NumConformers() - 1);
if (!pConv->IsOption("b",OBConversion::INOPTIONS))
mol.ConnectTheDots();
if (!pConv->IsOption("s",OBConversion::INOPTIONS) && !pConv->IsOption("b",OBConversion::INOPTIONS))
mol.PerceiveBondOrders();
// Attach unit cell translation vectors if found
if (numTranslationVectors > 0) {
OBUnitCell* uc = new OBUnitCell;
uc->SetData(acell[0] * translationVectors[0], acell[1] * translationVectors[1], acell[2] * translationVectors[2]);
uc->SetOrigin(fileformatInput);
if (symmetryCode)
uc->SetSpaceGroup(symmetryCode);
mol.SetData(uc);
}
mol.SetTitle(title);
return(true);
}
bool JaguarOutputFormat::ReadMolecule(OBBase* pOb, OBConversion* pConv)
{
OBMol* pmol = pOb->CastAndClear<OBMol>();
if(pmol==NULL) return false;
//Define some references so we can use the old parameter names
istream &ifs = *pConv->GetInStream();
OBMol &mol = *pmol;
const char* title = pConv->GetTitle();
char buffer[BUFF_SIZE];
string str,str1;
double x,y,z;
unsigned int i;
OBAtom *atom;
vector<string> vs;
mol.BeginModify();
while (ifs.getline(buffer,BUFF_SIZE))
{
if (strstr(buffer,"Input geometry:") != NULL
|| strstr(buffer,"symmetrized geometry:") != NULL
|| strstr(buffer,"new geometry:") != NULL
|| strstr(buffer,"final geometry:") != NULL)
{
// mol.EndModify();
mol.Clear();
mol.BeginModify();
ifs.getline(buffer,BUFF_SIZE); // angstroms
ifs.getline(buffer,BUFF_SIZE); // column headings
ifs.getline(buffer,BUFF_SIZE);
tokenize(vs,buffer);
while (vs.size() == 4)
{
atom = mol.NewAtom();
str = vs[0]; // Separate out the Symbol# into just Symbol ...
for (i = 0;i < str.size();i++)
if (isdigit(str[i]))
str[i] = '\0';
atom->SetAtomicNum(etab.GetAtomicNum(str.c_str()));
x = atof((char*)vs[1].c_str());
y = atof((char*)vs[2].c_str());
z = atof((char*)vs[3].c_str());
atom->SetVector(x,y,z);
if (!ifs.getline(buffer,BUFF_SIZE)) break;
tokenize(vs,buffer);
}
}
if (strstr(buffer, "Atomic charges from electrostatic potential") != NULL)
{
mol.SetAutomaticPartialCharge(false);
unsigned int chgcount=0;
while (chgcount<mol.NumAtoms())
{
ifs.getline(buffer,BUFF_SIZE); // blank line
ifs.getline(buffer,BUFF_SIZE); // header line
ifs.getline(buffer,BUFF_SIZE); // data line
tokenize(vs,buffer);
for (vector<string>::size_type icount=1;icount<vs.size();++icount)
{
chgcount=chgcount+1;
mol.GetAtom(chgcount)->SetPartialCharge(atof((char*)vs[icount].c_str()));
}
}
}
else if(strstr(buffer,"Dipole Moments (Debye)") != NULL)
{
ifs.getline(buffer,BUFF_SIZE); // actual components X ### Y #### Z ###
tokenize(vs,buffer);
if (vs.size() >= 8)
{
OBVectorData *dipoleMoment = new OBVectorData;
dipoleMoment->SetAttribute("Dipole Moment");
double x, y, z;
x = atof(vs[1].c_str());
y = atof(vs[3].c_str());
z = atof(vs[5].c_str());
dipoleMoment->SetData(x, y, z);
dipoleMoment->SetOrigin(fileformatInput);
mol.SetData(dipoleMoment);
}
if (!ifs.getline(buffer,BUFF_SIZE)) break;
}
}
if (!pConv->IsOption("b",OBConversion::INOPTIONS))
mol.ConnectTheDots();
if (!pConv->IsOption("s",OBConversion::INOPTIONS)
&& !pConv->IsOption("b",OBConversion::INOPTIONS))
mol.PerceiveBondOrders();
mol.EndModify();
mol.SetTitle(title);
return(true);
}
bool BGFFormat::ReadMolecule(OBBase* pOb, OBConversion* pConv)
{
OBMol* pmol = pOb->CastAndClear<OBMol>();
if(pmol==NULL)
return false;
//Define some references so we can use the old parameter names
istream &ifs = *pConv->GetInStream();
OBMol &mol = *pmol;
mol.SetTitle( pConv->GetTitle()); //default title is the filename
mol.BeginModify();
char buffer[BUFF_SIZE];
char tmp[16],tmptyp[16];
vector<string> vs;
while (ifs.getline(buffer,BUFF_SIZE)) {
if (EQn(buffer,"CRYSTX",6)) {
// Parse unit cell
tokenize(vs,buffer," \n\t,");
if (vs.size() != 7)
continue; // something strange
double A, B, C, Alpha, Beta, Gamma;
A = atof(vs[1].c_str());
B = atof(vs[2].c_str());
C = atof(vs[3].c_str());
Alpha = atof(vs[4].c_str());
Beta = atof(vs[5].c_str());
Gamma = atof(vs[6].c_str());
OBUnitCell *uc = new OBUnitCell;
uc->SetOrigin(fileformatInput);
uc->SetData(A, B, C, Alpha, Beta, Gamma);
mol.SetData(uc);
} else if (EQn(buffer,"FORMAT",6))
break;
}
ttab.SetFromType("DRE");
ttab.SetToType("INT");
OBAtom *atom;
double x,y,z,chrg;
for (;;)
{
if (!ifs.getline(buffer,BUFF_SIZE))
break;
if (EQn(buffer,"FORMAT",6))
break;
sscanf(buffer,"%*s %*s %*s %*s %*s %*s %lf %lf %lf %15s %*s %*s %lf",
&x,&y,&z,
tmptyp,
&chrg);
atom = mol.NewAtom();
ttab.Translate(tmp,tmptyp);
atom->SetType(tmp);
CleanAtomType(tmptyp);
atom->SetAtomicNum(etab.GetAtomicNum(tmptyp));
atom->SetVector(x,y,z);
}
unsigned int i;
vector<int> vtmp;
vector<vector<int> > vcon;
vector<vector<int> > vord;
for (i = 0; i < mol.NumAtoms();i++)
{
vcon.push_back(vtmp);
vord.push_back(vtmp);
}
unsigned int bgn;
for (;;)
{
if (!ifs.getline(buffer,BUFF_SIZE) || EQn(buffer,"END",3))
break;
tokenize(vs,buffer);
if (vs.empty() || vs.size() < 3 || vs.size() > 10)
continue;
if (EQn(buffer,"CONECT",6))
{
bgn = atoi((char*)vs[1].c_str()) - 1;
if (bgn < 1 || bgn > mol.NumAtoms())
continue;
for (i = 2;i < vs.size();i++)
{
vcon[bgn].push_back(atoi((char*)vs[i].c_str()));
vord[bgn].push_back(1);
}
}
else
if (EQn(buffer,"ORDER",5))
{
bgn = atoi((char*)vs[1].c_str()) - 1;
if (bgn < 1 || bgn > mol.NumAtoms())
continue;
if (vs.size() > vord[bgn].size()+2)
continue;
for (i = 2;i < vs.size();i++)
vord[bgn][i-2] = atoi((char*)vs[i].c_str());
}
}
unsigned int j;
for (i = 1;i <= mol.NumAtoms();i++)
if (!vcon[i - 1].empty())
for (j = 0;j < vcon[i - 1].size();j++)
{
mol.AddBond(i,vcon[i - 1][j],vord[i - 1][j]);
}
//load up the next line after the END marker
ifs.getline(buffer,BUFF_SIZE);
mol.EndModify();
return(true);
}
bool XSFFormat::ReadMolecule(OBBase* pOb, OBConversion* pConv)
{
OBMol* pmol = pOb->CastAndClear<OBMol>();
if(pmol==NULL)
return false;
//Define some references so we can use the old parameter names
istream &ifs = *pConv->GetInStream();
OBMol &mol = *pmol;
const char* title = pConv->GetTitle();
char buffer[BUFF_SIZE];
string str;
double x,y,z;
OBAtom *atom;
vector3 translationVectors[3];
int numTranslationVectors = 0;
vector<string> vs;
vector<vector3> atomPositions;
bool createdAtoms = false;
int atomicNum;
mol.BeginModify();
while (ifs.getline(buffer, BUFF_SIZE))
{
if (buffer[0] == '#')
continue; // comment
if (strstr(buffer, "ATOMS") != NULL) {
// Minimum of 4 columns -- AtNum, x, y, z (forces)
// where AtNum stands for atomic number (or symbol), while X Y Z are
ifs.getline(buffer, BUFF_SIZE);
tokenize(vs, buffer);
while (vs.size() >= 4) {
if (!createdAtoms) {
atom = mol.NewAtom();
//set atomic number
atomicNum = OBElements::GetAtomicNum(vs[0].c_str());
if (atomicNum == 0) {
atomicNum = atoi(vs[0].c_str());
}
atom->SetAtomicNum(atomicNum);
}
x = atof((char*)vs[1].c_str());
y = atof((char*)vs[2].c_str());
z = atof((char*)vs[3].c_str());
atomPositions.push_back(vector3(x, y, z)); // we may have a movie or animation
ifs.getline(buffer, BUFF_SIZE);
tokenize(vs, buffer);
}
createdAtoms = true; // don't run NewAtom() anymore
}
else if ( strstr(buffer, "PRIMVEC")
|| strstr(buffer, "CONVVEC") ) {
// translation vectors
numTranslationVectors = 0; // if we have an animation
while (numTranslationVectors < 3 && ifs.getline(buffer,BUFF_SIZE)) {
tokenize(vs,buffer); // we really need to check that it's 3 entries only
if (vs.size() < 3) return false; // timvdm 18/06/2008
x = atof((char*)vs[0].c_str());
y = atof((char*)vs[1].c_str());
z = atof((char*)vs[2].c_str());
translationVectors[numTranslationVectors++].Set(x, y, z);
}
}
else if (strstr(buffer, "PRIMCOORD") != NULL) {
// read the coordinates
ifs.getline(buffer, BUFF_SIZE);
tokenize(vs, buffer);
if (vs.size() < 2) return false;
int numAtoms = atoi(vs[0].c_str());
for (int a = 0; a < numAtoms; ++a) {
if (!ifs.getline(buffer,BUFF_SIZE))
break;
tokenize(vs,buffer);
if (vs.size() < 4)
break;
if (!createdAtoms) {
atom = mol.NewAtom();
//set atomic number
atomicNum = OBElements::GetAtomicNum(vs[0].c_str());
if (atomicNum == 0) {
atomicNum = atoi(vs[0].c_str());
}
atom->SetAtomicNum(atomicNum);
}
x = atof((char*)vs[1].c_str());
y = atof((char*)vs[2].c_str());
z = atof((char*)vs[3].c_str());
atomPositions.push_back(vector3(x, y, z));
}
}
}
mol.EndModify();
int natom = mol.NumAtoms();
int numConformers = atomPositions.size() / natom;
for (int i = 0; i < numConformers; ++i) {
double *coordinates = new double[natom * 3];
for (int j = 0; j < natom; ++j) {
vector3 currentPosition = atomPositions[i*natom + j];
coordinates[j*3] = currentPosition.x();
coordinates[j*3 + 1] = currentPosition.y();
coordinates[j*3 + 2] = currentPosition.z();
}
mol.AddConformer(coordinates);
}
// Delete first conformer, created by EndModify, bunch of 0s
mol.DeleteConformer(0);
// Set geometry to last one
mol.SetConformer(mol.NumConformers() - 1);
if (!pConv->IsOption("b",OBConversion::INOPTIONS))
mol.ConnectTheDots();
if (!pConv->IsOption("s",OBConversion::INOPTIONS) && !pConv->IsOption("b",OBConversion::INOPTIONS))
mol.PerceiveBondOrders();
// Add final properties
mol.SetTitle(title);
if (numTranslationVectors == 3) {
OBUnitCell *uc = new OBUnitCell;
uc->SetOrigin(fileformatInput);
uc->SetData(translationVectors[0],
translationVectors[1],
translationVectors[2]);
mol.SetData(uc);
}
return(true);
}
bool CARFormat::ReadMolecule(OBBase* pOb, OBConversion* pConv)
{
OBMol* pmol = pOb->CastAndClear<OBMol>();
if(pmol==NULL)
return false;
//Define some references so we can use the old parameter names
istream &ifs = *pConv->GetInStream();
OBMol &mol = *pmol;
const char* title = pConv->GetTitle();
bool hasPartialCharges = false;
char buffer[BUFF_SIZE];
string str;
double x,y,z;
OBAtom *atom;
vector<string> vs;
mol.BeginModify();
while (ifs.getline(buffer,BUFF_SIZE))
{
if(strstr(buffer,"end") != NULL)
{
if (mol.NumAtoms() > 0) // we've already read in a molecule, so exit
break;
// else, we hit the end of the previous molecular system
// (in a multimolecule file)
ifs.getline(buffer,BUFF_SIZE); // title
ifs.getline(buffer,BUFF_SIZE); // DATE
}
if (strncmp(buffer, "!BIOSYM", 7) == 0)
{
continue;
}
if(strstr(buffer,"PBC") != NULL)
{
if(strstr(buffer,"ON") != NULL)
{
ifs.getline(buffer,BUFF_SIZE); // title
ifs.getline(buffer,BUFF_SIZE); // DATE
ifs.getline(buffer,BUFF_SIZE); // PBC a b c alpha beta gamma SG
// parse cell parameters
tokenize(vs,buffer);
if (vs.size() == 8)
{
//parse cell values
double A,B,C,Alpha,Beta,Gamma;
A = atof((char*)vs[1].c_str());
B = atof((char*)vs[2].c_str());
C = atof((char*)vs[3].c_str());
Alpha = atof((char*)vs[4].c_str());
Beta = atof((char*)vs[5].c_str());
Gamma = atof((char*)vs[6].c_str());
OBUnitCell *uc = new OBUnitCell;
uc->SetOrigin(fileformatInput);
uc->SetData(A, B, C, Alpha, Beta, Gamma);
uc->SetSpaceGroup(vs[7]);
mol.SetData(uc);
}
}
else // PBC=OFF
{
ifs.getline(buffer,BUFF_SIZE); // title
ifs.getline(buffer,BUFF_SIZE); // !DATE
}
continue;
} // PBC
// reading real data!
tokenize(vs,buffer);
if (vs.size() < 8) {
break;
}
atom = mol.NewAtom();
atom->SetAtomicNum(etab.GetAtomicNum(vs[7].c_str()));
x = atof((char*)vs[1].c_str());
y = atof((char*)vs[2].c_str());
z = atof((char*)vs[3].c_str());
atom->SetVector(x,y,z);
// vs[0] contains atom label
// vs[4] contains "type of residue containing atom"
// vs[5] contains "residue sequence name"
// vs[6] contains "potential type of atom"
if (vs.size() == 9)
{
atom->SetPartialCharge(atof((char*)vs[8].c_str()));
hasPartialCharges = true;
}
}
if (!pConv->IsOption("b",OBConversion::INOPTIONS))
mol.ConnectTheDots();
if (!pConv->IsOption("s",OBConversion::INOPTIONS) && !pConv->IsOption("b",OBConversion::INOPTIONS))
mol.PerceiveBondOrders();
mol.EndModify();
if (hasPartialCharges)
mol.SetPartialChargesPerceived();
mol.SetTitle(title);
return(true);
}
static bool parseAtomRecord(char *buffer, OBMol &mol,int /*chainNum*/)
/* ATOMFORMAT "(i5,1x,a4,a1,a3,1x,a1,i4,a1,3x,3f8.3,2f6.2,a2,a2)" */
{
string sbuf = &buffer[6];
if (sbuf.size() < 48)
return(false);
bool hetatm = (EQn(buffer,"HETATM",6)) ? true : false;
bool elementFound = false; // true if correct element found in col 77-78
/* serial number */
string serno = sbuf.substr(0,5);
/* atom name */
string atmid = sbuf.substr(6,4);
/* chain */
char chain = sbuf.substr(15,1)[0];
/* element */
string element = " ";
if (sbuf.size() > 71)
{
element = sbuf.substr(70,2);
if (isalpha(element[1]))
{
if (element[0] == ' ')
{
element.erase(0, 1);
elementFound = true;
}
else if (isalpha(element[0]))
{
elementFound = true;
}
}
}
if (!elementFound)
{
stringstream errorMsg;
errorMsg << "WARNING: Problems reading a PDB file\n"
<< " Problems reading a HETATM or ATOM record.\n"
<< " According to the PDB specification,\n"
<< " columns 77-78 should contain the element symbol of an atom.\n"
<< " but OpenBabel found '" << element << "' (atom " << mol.NumAtoms()+1 << ")";
obErrorLog.ThrowError(__FUNCTION__, errorMsg.str(), obWarning);
}
// charge - optional
string scharge;
if (sbuf.size() > 73)
{
scharge = sbuf.substr(72,2);
}
//trim spaces on the right and left sides
while (!atmid.empty() && atmid[0] == ' ')
atmid = atmid.erase(0, 1);
while (!atmid.empty() && atmid[atmid.size()-1] == ' ')
atmid = atmid.substr(0,atmid.size()-1);
/* residue name */
string resname = sbuf.substr(11,3);
if (resname == " ")
resname = "UNK";
else
{
while (!resname.empty() && resname[0] == ' ')
resname = resname.substr(1,resname.size()-1);
while (!resname.empty() && resname[resname.size()-1] == ' ')
resname = resname.substr(0,resname.size()-1);
}
string type;
if (!elementFound) {
// OK, we have to fall back to determining the element from the atom type
// This is unreliable, but there's no other choice
if (EQn(buffer,"ATOM",4)) {
type = atmid.substr(0,2);
if (isdigit(type[0])) {
// sometimes non-standard files have, e.g 11HH
if (!isdigit(type[1])) type = atmid.substr(1,1);
else type = atmid.substr(2,1);
} else if ((sbuf[6] == ' ' &&
strncasecmp(type.c_str(), "Zn", 2) != 0 &&
strncasecmp(type.c_str(), "Fe", 2) != 0) ||
isdigit(type[1])) //type[1] is digit in Platon
type = atmid.substr(0,1); // one-character element
if (resname.substr(0,2) == "AS" || resname[0] == 'N') {
if (atmid == "AD1")
type = "O";
if (atmid == "AD2")
type = "N";
}
if (resname.substr(0,3) == "HIS" || resname[0] == 'H') {
if (atmid == "AD1" || atmid == "AE2")
type = "N";
if (atmid == "AE1" || atmid == "AD2")
type = "C";
}
if (resname.substr(0,2) == "GL" || resname[0] == 'Q') {
if (atmid == "AE1")
type = "O";
if (atmid == "AE2")
type = "N";
}
// fix: #2002557
if (atmid[0] == 'H' &&
(atmid[1] == 'D' || atmid[1] == 'E' ||
atmid[1] == 'G' || atmid[1] == 'H')) // HD, HE, HG, HH, ..
type = "H";
} else { //must be hetatm record
if (isalpha(element[1]) && (isalpha(element[0]) || (element[0] == ' '))) {
if (isalpha(element[0]))
type = element.substr(0,2);
else
type = element.substr(1,1);
if (type.size() == 2)
type[1] = tolower(type[1]);
} else { // no element column to use
if (isalpha(atmid[0])) {
if (atmid.size() > 2 && (atmid[2] == '\0' || atmid[2] == ' '))
type = atmid.substr(0,2);
else if (atmid[0] == 'A') // alpha prefix
type = atmid.substr(1, atmid.size() - 1);
else
type = atmid.substr(0,1);
} else if (atmid[0] == ' ')
type = atmid.substr(1,1); // one char element
else
type = atmid.substr(1,2);
// Some cleanup steps
if (atmid == resname) {
type = atmid;
if (type.size() == 2)
type[1] = tolower(type[1]);
} else
if (resname == "ADR" || resname == "COA" || resname == "FAD" ||
resname == "GPG" || resname == "NAD" || resname == "NAL" ||
resname == "NDP" || resname == "ABA") {
if (type.size() > 1)
type = type.substr(0,1);
//type.erase(1,type.size()-1);
} else // other residues
if (isdigit(type[0])){
type = type.substr(1,1);
}
else
if (type.size() > 1 && isdigit(type[1]))
type = type.substr(0,1);
else
if (type.size() > 1 && isalpha(type[1])) {
if (type[0] == 'O' && type[1] == 'H')
type = type.substr(0,1); // no "Oh" element (e.g. 1MBN)
else if(isupper(type[1])) {
type[1] = tolower(type[1]);
}
}
}
} // HETATM records
} // no element column to use
OBAtom atom;
/* X, Y, Z */
string xstr = sbuf.substr(24,8);
string ystr = sbuf.substr(32,8);
string zstr = sbuf.substr(40,8);
vector3 v(atof(xstr.c_str()),atof(ystr.c_str()),atof(zstr.c_str()));
atom.SetVector(v);
atom.ForceImplH();
// useful for debugging unknown atom types (e.g., PR#1577238)
// cout << mol.NumAtoms() + 1 << " : '" << element << "'" << " " << etab.GetAtomicNum(element.c_str()) << endl;
if (elementFound)
atom.SetAtomicNum(etab.GetAtomicNum(element.c_str()));
else // use our old-style guess from athe atom type
atom.SetAtomicNum(etab.GetAtomicNum(type.c_str()));
if ( (! scharge.empty()) && " " != scharge )
{
if ( isdigit(scharge[0]) && ('+' == scharge[1] || '-' == scharge[1]) )
{
const char reorderCharge[3] = { scharge[1], scharge[0], '\0' };
const int charge = atoi(reorderCharge);
atom.SetFormalCharge(charge);
}
else
{
stringstream errorMsg;
errorMsg << "WARNING: Problems reading a PDB file\n"
<< " Problems reading a HETATM or ATOM record.\n"
<< " According to the PDB specification,\n"
<< " columns 79-80 should contain charge of the atom\n"
<< " but OpenBabel found '" << scharge << "' (atom " << mol.NumAtoms()+1 << ").";
obErrorLog.ThrowError(__FUNCTION__, errorMsg.str(), obWarning);
}
}
else {
atom.SetFormalCharge(0);
}
/* residue sequence number */
string resnum = sbuf.substr(16,4);
OBResidue *res = (mol.NumResidues() > 0) ? mol.GetResidue(mol.NumResidues()-1) : NULL;
if (res == NULL
|| res->GetName() != resname
|| res->GetNumString() != resnum
|| res->GetChain() != chain)
{
vector<OBResidue*>::iterator ri;
for (res = mol.BeginResidue(ri) ; res ; res = mol.NextResidue(ri))
if (res->GetName() == resname
&& res->GetNumString() == resnum
&& static_cast<int>(res->GetChain()) == chain)
break;
if (res == NULL) {
res = mol.NewResidue();
res->SetChain(chain);
res->SetName(resname);
res->SetNum(resnum);
}
}
if (!mol.AddAtom(atom))
return(false);
else {
OBAtom *atom = mol.GetAtom(mol.NumAtoms());
res->AddAtom(atom);
res->SetSerialNum(atom, atoi(serno.c_str()));
res->SetAtomID(atom, sbuf.substr(6,4));
res->SetHetAtom(atom, hetatm);
return(true);
}
} // end reading atom records
bool TinkerFormat::ReadMolecule(OBBase* pOb, OBConversion* pConv)
{
OBMol* pmol = pOb->CastAndClear<OBMol>();
if(pmol==NULL)
return false;
//Define some references so we can use the old parameter names
istream &ifs = *pConv->GetInStream();
OBMol &mol = *pmol;
const char* title = pConv->GetTitle();
int natoms;
char buffer[BUFF_SIZE];
vector<string> vs;
ifs.getline(buffer, BUFF_SIZE);
tokenize(vs,buffer);
if (vs.size() < 2)
return false;
natoms = atoi(vs[0].c_str());
// title is 2nd token (usually add tokens for the atom types)
mol.SetTitle(vs[1]);
mol.ReserveAtoms(natoms);
mol.BeginModify();
string str;
double x,y,z;
OBAtom *atom;
for (int i = 1; i <= natoms; ++i)
{
if (!ifs.getline(buffer,BUFF_SIZE))
return(false);
tokenize(vs,buffer);
// e.g. "2 C 2.476285 0.121331 -0.001070 2 1 3 14"
if (vs.size() < 5)
return(false);
atom = mol.NewAtom();
x = atof((char*)vs[2].c_str());
y = atof((char*)vs[3].c_str());
z = atof((char*)vs[4].c_str());
atom->SetVector(x,y,z); //set coordinates
//set atomic number
atom->SetAtomicNum(OBElements::GetAtomicNum(vs[1].c_str()));
// add bonding
if (vs.size() > 6)
for (unsigned int j = 6; j < vs.size(); ++j)
mol.AddBond(mol.NumAtoms(), atoi((char *)vs[j].c_str()), 1); // we don't know the bond order
}
if (!pConv->IsOption("s",OBConversion::INOPTIONS))
mol.PerceiveBondOrders();
// clean out remaining blank lines
std::streampos ipos;
do
{
ipos = ifs.tellg();
ifs.getline(buffer,BUFF_SIZE);
}
while(strlen(buffer) == 0 && !ifs.eof() );
ifs.seekg(ipos);
mol.EndModify();
mol.SetTitle(title);
return(true);
}
bool ChemDrawXMLFormat::DoElement(const string& name)
{
string buf;
if(name=="fragment")
{
//This is the start of the molecule we are extracting and it will
//be put into the OBMol* _pmol declared in the parent class.
//initialise everything
_tempAtom.Clear();
atoms.clear();
_pmol->SetDimension(2);
_pmol->BeginModify();
buf = _pxmlConv->GetAttribute("id");
if (buf.length())
{
_pmol->SetTitle(buf);
}
}
else if(name=="n")
{
EnsureEndElement();
buf = _pxmlConv->GetAttribute("Type");
if (buf.length())
{
if (buf != "Unspecified" && buf != "Element")
{
cerr << "CDXML Format: Node type \"" << buf <<
"\" is not currently supported." << endl;
return false; // FIXME: use as many types as possible
}
}
_tempAtom.SetAtomicNum(6); // default is carbon
buf = _pxmlConv->GetAttribute("id");
if (buf.length())
_tempAtom.SetIdx(atoi(buf.c_str()));
buf = _pxmlConv->GetAttribute("Element");
if (buf.length())
_tempAtom.SetAtomicNum(atoi(buf.c_str()));
buf = _pxmlConv->GetAttribute("p"); // coords
if (buf.length())
{
double x = 0., y = 0.;
sscanf(buf.c_str(), "%lf %lf", &x, &y);
_tempAtom.SetVector(x, y, 0.);
}
buf = _pxmlConv->GetAttribute("Charge");
if (buf.length())
_tempAtom.SetFormalCharge(atoi(buf.c_str()));
}
else if(name=="b")
{
EnsureEndElement();
bool invert_ends = false;
Begin = End = Flag = 0;
buf = _pxmlConv->GetAttribute("Order");
if (buf.length())
Order = atoi(buf.c_str());
else
Order = 1; //default value
buf = _pxmlConv->GetAttribute("Display");
if (buf.length())
{
if (buf == "WedgeEnd")
{
invert_ends = true;
Flag = OB_HASH_BOND;
}
else if (buf == "WedgeBegin")
{
Flag = OB_HASH_BOND;
}
else if (buf == "Hash" ||buf == "WedgedHashBegin")
{
Flag = OB_WEDGE_BOND;
}
else if (buf == "WedgedHashEnd")
{
invert_ends = true;
Flag = OB_WEDGE_BOND;
}
}
buf = _pxmlConv->GetAttribute("B");
if (buf.length())
{
if (invert_ends)
End = atoms[atoi(buf.c_str())];
else
Begin = atoms[atoi(buf.c_str())];
}
buf = _pxmlConv->GetAttribute("E");
if (buf.length())
{
if (invert_ends)
Begin = atoms[atoi(buf.c_str())];
else
End = atoms[atoi(buf.c_str())];
}
}
/*
// Forget that, the fragment, aka molecule, is in another XML hierachy tree than the data.
// Parsing has already stopped before ever getting to this point
else if(name=="tags")
{
buf = _pxmlConv->GetAttribute("ID");
if (buf.length())
{
}
}
else if(name=="tableCell")
{
buf = _pxmlConv->GetAttribute("value");
if (buf.length())
{
}
}
*/
return true;
}
bool ACRFormat::ReadMolecule(OBBase* pOb, OBConversion* pConv)
{
OBMol* pmol = pOb->CastAndClear<OBMol>();
if(pmol==NULL)
return false;
istream& ifs = *pConv->GetInStream();
pmol->BeginModify();
/** Parse the input stream and use the OpenBabel API to populate the OBMol **/
int id;
char buf[BUFF_SIZE];
int atoms, bonds, tmp;
float scale, dtmp;
bool atom_input = false, bond_input = false;
string type;
//int from, to;
double X,Y,Z;
vector<string> vs;
// read in one at a time
/* WARNING: Atom id starts from zero in Carine; not so in openbabel.
* Solution: increment atom id's */
while (true) {
ifs.getline(buf, BUFF_SIZE);
if (ifs.eof()) {
break;
}
if (sscanf(buf, "General Scale=%f\n", &dtmp)) {
scale = dtmp;
continue;
} else if (sscanf(buf, "Number of Atoms in Crystal=%d\n", &tmp)) {
atoms = tmp;
atom_input = true;
// read table column names
ifs.getline(buf, BUFF_SIZE);
continue;
} else if (sscanf(buf, "Number of Links in Crystal=%d\n", &tmp)) {
atom_input = false;
bond_input = true;
bonds = tmp;
// read table column names
ifs.getline(buf, BUFF_SIZE);
continue;
} else if ( '#' == buf[0] || '\r' == buf[0] || '\n' == buf[0] ) {
// between sections, in both windows and unix.
continue;
}
tokenize(vs, buf, " \t\r\n");
if (atom_input) {
if (vs.size() < 9) return false; // timvdm 18/06/2008
id = atoi((char*)vs[0].c_str()) + 1; // see warning above
type = vs[1];
X = atof((char*)vs[6].c_str())/scale;
Y = atof((char*)vs[7].c_str())/scale;
Z = atof((char*)vs[8].c_str())/scale;
OBAtom* a = pmol->NewAtom();
if (*(type.c_str()) != '*')
a->SetAtomicNum(etab.GetAtomicNum(type.c_str()));
a->SetVector(X,Y,Z);
} else if (bond_input) {
if (vs.size() < 2) return false; // timvdm 18/06/2008
// add to pmol
if (!pmol->AddBond(atoi((char*)vs[0].c_str()) + 1, atoi((char*)vs[1].c_str()) + 1,
1 /* bond order not specified in Carine, use PerceiveBondOrder later */))
{
obErrorLog.ThrowError(__FUNCTION__, "addition of bond between " + vs[0] + " and " + vs[1] + " failed", obError);
return false;
}
}
}
/* got sanity? */
if ( pmol->NumBonds() != bonds ) {
// then we read a different number of bonds than those promised.
obErrorLog.ThrowError(__FUNCTION__, "Number of bonds read does not match the number promised", obError);
return false;
} else if ( pmol->NumAtoms() != atoms ) {
obErrorLog.ThrowError(__FUNCTION__, "Number of atoms read does not match the number promised", obError);
return false;
}
pmol->PerceiveBondOrders();
pmol->EndModify();
return true;
}
bool ReadViewMol(istream &ifs,OBMol &mol,const char *title)
{
char buffer[BUFF_SIZE];
OBAtom *atom;
double x,y,z, border;
double factor = 1.0;
int bgn, end, order;
vector<string> vs;
bool foundTitle = false;
bool foundBonds = false;
mol.BeginModify();
while (ifs.getline(buffer,BUFF_SIZE))
{
if (strstr(buffer,"$title") != NULL)
{
if (!ifs.getline(buffer,BUFF_SIZE)) return (false);
mol.SetTitle(buffer);
foundTitle = true;
}
else if (strstr(buffer,"$coord") != NULL)
{
tokenize(vs,buffer);
if (vs.size() == 2)
factor = atof((char*)vs[1].c_str()); // conversion to angstrom
while (ifs.getline(buffer,BUFF_SIZE))
{
if (buffer[0] == '$') break;
tokenize(vs,buffer);
if (vs.size() != 4) return(false);
atom = mol.NewAtom();
x = atof((char*)vs[0].c_str()) * factor;
y = atof((char*)vs[1].c_str()) * factor;
z = atof((char*)vs[2].c_str()) * factor;
atom->SetVector(x,y,z); //set coordinates
atom->SetAtomicNum(etab.GetAtomicNum(vs[3].c_str()));
}
}
else if (strstr(buffer,"$bonds") != NULL)
{
foundBonds = true;
while (ifs.getline(buffer,BUFF_SIZE))
{
if (buffer[0] == '$') break;
sscanf(buffer,"%d %d %lf",&bgn,&end, &border);
if (border > 1.0)
order = int(border);
else
order = 1;
mol.AddBond(bgn+1,end+1,order);
}
}
else if (strstr(buffer,"$end") != NULL)
break;
} // while
mol.EndModify();
if (!foundTitle)
mol.SetTitle(title);
if (!foundBonds)
{
mol.ConnectTheDots();
mol.PerceiveBondOrders();
}
return(true);
}
bool ThermoFormat::ReadMolecule(OBBase* pOb, OBConversion* pConv)
{
OBMol* pmol = pOb->CastAndClear<OBMol>();
if(!pmol)
return false;
bool stopOnEnd = pConv->IsOption("e",OBConversion::INOPTIONS)!=NULL;
pmol->SetDimension(0);
OBNasaThermoData* pND = new OBNasaThermoData; //to store rate constant data
pND->SetOrigin(fileformatInput);
pmol->SetData(pND);
istream &ifs = *pConv->GetInStream();
double DefaultMidT = 1500;
char ln[BUFF_SIZE];
unsigned int i;
//find line with 1 in col 80
do
{
if(!ifs.getline(ln,BUFF_SIZE) || stopOnEnd && !strncasecmp(ln,"END",3))
return false;
}while(ln[79]!='1');
char phase, nam[25], dum[7], elname[3];
elname[2]=0;
int elnum;
double Coeff[14];
sscanf(ln,"%18s%6s",nam,dum);
pmol->SetTitle(nam);
char* p=ln+24;
if(ln[80]=='&')
{
//Reaction Design extension
p+=20;
string line;
if(!getline(ifs,line))return false;
vector<string> toks;
tokenize(toks,line," \t\n\r");
for(i=0;i<toks.size();i+=2)
{
OBAtom atom;
atom.SetAtomicNum(etab.GetAtomicNum(toks[i].c_str()));
elnum = atoi(toks[i+1].c_str());
atom.ForceNoH();
for(;elnum>0;--elnum)
pmol->AddAtom(atom);
}
}
else
{
for(i=0;i<4;i++,p+=5)
{
char snum[4]={0,0,0,0};//Was problem with F 10 0 reading as ten
sscanf(p,"%c%c%c%c%c",elname,elname+1,snum,snum+1,snum+2);
elnum=atoi(snum);
if(elname[0]!=' ' && elname[0]!='0')
{
if(elname[1]==' ')
elname[1]=0;
OBAtom atom;
atom.SetAtomicNum(etab.GetAtomicNum(elname));
atom.ForceNoH();
for(;elnum>0;--elnum)
pmol->AddAtom(atom);
}
}
}
double LoT, HiT, MidT=0;
/* int nc = */sscanf(p,"%c%10lf%10lf10%lf",&phase, &LoT, &HiT, &MidT);
pND->SetPhase(phase);
pND->SetLoT(LoT);
pND->SetHiT(HiT);
if(MidT>HiT || MidT<LoT)
MidT=DefaultMidT;
pND->SetMidT(MidT);
if (!ifs.getline(ln, BUFF_SIZE)) return false;
p=ln;
for(i=0;i<5;i++,p+=15)
sscanf(p,"%15lf",&Coeff[i]);
if (!ifs.getline(ln, BUFF_SIZE)) return false;
p=ln;
for(i=5;i<10;i++,p+=15)
sscanf(p,"%15lf",&Coeff[i]);
if (!ifs.getline(ln, BUFF_SIZE)) return false;
p=ln;
for(i=10;i<14;i++,p+=15)
sscanf(p,"%15lf",&Coeff[i]);
for(i=0;i<14;++i)
pND->SetCoeff(i, Coeff[i]);
pmol->AssignSpinMultiplicity();
return true;
}
bool MSIFormat::ReadMolecule(OBBase* pOb, OBConversion* pConv)
{
OBMol* pmol = pOb->CastAndClear<OBMol>();
if(pmol==NULL)
return false;
//Define some references so we can use the old parameter names
istream &ifs = *pConv->GetInStream();
OBMol &mol = *pmol;
const char* title = pConv->GetTitle();
char buffer[BUFF_SIZE];
stringstream errorMsg;
if (!ifs)
return false; // we're attempting to read past the end of the file
if (!ifs.getline(buffer,BUFF_SIZE))
{
obErrorLog.ThrowError(__FUNCTION__,
"Problems reading an MSI file: Cannot read the first line.", obWarning);
return(false);
}
if (!EQn(buffer, "# MSI CERIUS2 DataModel File", 28))
{
obErrorLog.ThrowError(__FUNCTION__,
"Problems reading an MSI file: The first line must contain the MSI header.", obWarning);
return(false);
}
// "records" start with
// (1 Model
// ....
// and end with
// ....
// )
unsigned int openParens = 0; // the count of "open parentheses" tags
unsigned int startBondAtom, endBondAtom, bondOrder;
bool atomRecord = false;
bool bondRecord = false;
OBAtom *atom;
// OBBond *bond;
vector<string> vs;
const SpaceGroup *sg;
bool setSpaceGroup = false;
double x,y,z;
vector3 translationVectors[3];
int numTranslationVectors = 0;
mol.BeginModify();
while (ifs.getline(buffer,BUFF_SIZE))
{
// model record
if (strstr(buffer, "Model") != NULL) {
openParens++;
continue;
}
// atom record
if (!bondRecord && strstr(buffer, "Atom") != NULL) {
atomRecord = true;
openParens++;
continue;
}
if (strstr(buffer, "Bond") != NULL) {
bondRecord = true;
startBondAtom = endBondAtom = 0;
bondOrder = 1;
openParens++;
continue;
}
/* (A I PeriodicType 100)
(A D A3 (6.2380000000000004 0 0))
(A D B3 (0 6.9909999999999997 0))
(A D C3 (0 0 6.9960000000000004))
(A C SpaceGroup "63 5")
*/
if (strstr(buffer, "PeriodicType") != NULL) {
ifs.getline(buffer,BUFF_SIZE); // next line should be translation vector
tokenize(vs,buffer);
while (vs.size() == 6) {
x = atof((char*)vs[3].erase(0,1).c_str());
y = atof((char*)vs[4].c_str());
z = atof((char*)vs[5].c_str());
translationVectors[numTranslationVectors++].Set(x, y, z);
if (!ifs.getline(buffer,BUFF_SIZE))
break;
tokenize(vs,buffer);
}
}
if (strstr(buffer, "SpaceGroup") != NULL) {
tokenize(vs, buffer);
if (vs.size() != 5)
continue; // invalid space group
setSpaceGroup = true;
sg = SpaceGroup::GetSpaceGroup(vs[4]); // remove the initial " character
}
// atom information
if (atomRecord) {
if (strstr(buffer, "ACL") != NULL) {
tokenize(vs, buffer);
// size should be 5 -- need a test here
if (vs.size() != 5) return false; // timvdm 18/06/2008
vs[3].erase(0,1); // "6 => remove the first " character
int atomicNum = atoi(vs[3].c_str());
if (atomicNum == 0)
atomicNum = 1; // hydrogen ?
if (atomicNum <= 0 || atomicNum > etab.GetNumberOfElements())
continue;
// valid element, so create the atom
atom = mol.NewAtom();
atom->SetAtomicNum(atomicNum);
continue;
}
else if (strstr(buffer, "XYZ") != NULL) {
tokenize(vs, buffer);
// size should be 6 -- need a test here
if (vs.size() != 6) return false; // timvdm 18/06/2008
vs[3].erase(0,1); // remove ( character
vs[5].erase(vs[5].length()-2, 2); // remove trailing )) characters
atom->SetVector(atof(vs[3].c_str()),
atof(vs[4].c_str()),
atof(vs[5].c_str()));
continue;
}
} // end of atom records
// bond information
if (bondRecord) {
if (strstr(buffer, "Atom1") != NULL) {
tokenize(vs, buffer);
if (vs.size() < 4) return false; // timvdm 18/06/2008
vs[3].erase(vs[3].length()-1,1);
startBondAtom = atoi(vs[3].c_str());
continue;
}
else if (strstr(buffer, "Atom2") != NULL) {
tokenize(vs, buffer);
if (vs.size() < 4) return false; // timvdm 18/06/2008
vs[3].erase(vs[3].length()-1,1);
endBondAtom = atoi(vs[3].c_str());
continue;
}
else if (strstr(buffer, "Type") != NULL) {
tokenize(vs, buffer);
if (vs.size() < 4) return false; // timvdm 18/06/2008
vs[3].erase(vs[3].length()-1,1);
bondOrder = atoi(vs[3].c_str());
if (bondOrder == 4) // triple bond?
bondOrder = 3;
else if (bondOrder == 8) // aromatic?
bondOrder = 5;
else if (bondOrder != 2) // 1 OK, 2 OK, others unknown
bondOrder = 1;
continue;
}
}
// ending a "tag" -- a lone ")" on a line
if (strstr(buffer,")") != NULL && strstr(buffer, "(") == NULL) {
openParens--;
if (atomRecord) {
atomRecord = false;
}
if (bondRecord) {
// Bond records appear to be questionable
mol.AddBond(startBondAtom - 1, endBondAtom - 1, bondOrder);
bondRecord = false;
}
if (openParens == 0) {
ifs.getline(buffer, BUFF_SIZE);
break; // closed this molecule
}
}
}
mol.EndModify();
// clean out any remaining blank lines
while(ifs.peek() != EOF && ifs.good() &&
(ifs.peek() == '\n' || ifs.peek() == '\r'))
ifs.getline(buffer,BUFF_SIZE);
/*
if (!pConv->IsOption("b",OBConversion::INOPTIONS))
mol.ConnectTheDots();
if (!pConv->IsOption("s",OBConversion::INOPTIONS) && !pConv->IsOption("b",OBConversion::INOPTIONS))
mol.PerceiveBondOrders();
*/
if (numTranslationVectors > 0) {
OBUnitCell* uc = new OBUnitCell;
uc->SetData(translationVectors[0], translationVectors[1], translationVectors[2]);
uc->SetOrigin(fileformatInput);
if (setSpaceGroup) {
uc->SetSpaceGroup(sg);
}
mol.SetData(uc);
}
return(true);
}
bool AlchemyFormat::ReadMolecule(OBBase* pOb, OBConversion* pConv)
{
OBMol* pmol = pOb->CastAndClear<OBMol>();
if(pmol==NULL)
return false;
//Define some references so we can use the old parameter names
istream &ifs = *pConv->GetInStream();
OBMol &mol = *pmol;
const char* title = pConv->GetInFilename().c_str();
int i;
int natoms = 0, nbonds = 0;
char buffer[BUFF_SIZE];
ifs.getline(buffer,BUFF_SIZE);
sscanf(buffer," %d %*s %d", &natoms, &nbonds);
if (!natoms)
{
ifs.getline(buffer,BUFF_SIZE);
sscanf(buffer," %d %*s %d", &natoms, &nbonds);
if (!natoms)
return(false);
}
mol.ReserveAtoms(natoms);
mol.BeginModify();
ttab.SetFromType("ALC");
string str;
double x,y,z;
OBAtom *atom;
vector<string> vs;
for (i = 1; i <= natoms; i ++)
{
if (!ifs.getline(buffer,BUFF_SIZE))
return(false);
tokenize(vs,buffer);
if (vs.size() < 5)
return(false);
atom = mol.NewAtom();
x = atof((char*)vs[2].c_str());
y = atof((char*)vs[3].c_str());
z = atof((char*)vs[4].c_str());
atom->SetVector(x,y,z); //set coordinates
//set atomic number
ttab.SetToType("ATN");
ttab.Translate(str,vs[1]);
atom->SetAtomicNum(atoi(str.c_str()));
//set type
ttab.SetToType("INT");
ttab.Translate(str,vs[1]);
atom->SetType(str);
}
char bobuf[100];
string bostr;
int bgn,end,order;
for (i = 0; i < nbonds; i++)
{
if (!ifs.getline(buffer,BUFF_SIZE))
return(false);
sscanf(buffer," %*d%d%d%99s",&bgn,&end,bobuf);
bostr = bobuf;
order = 1;
if (bostr == "DOUBLE")
order = 2;
else if (bostr == "TRIPLE")
order = 3;
else if (bostr == "AROMATIC")
order = 5;
mol.AddBond(bgn,end,order);
}
// clean out remaining blank lines
while(ifs.peek() != EOF && ifs.good() &&
(ifs.peek() == '\n' || ifs.peek() == '\r'))
ifs.getline(buffer,BUFF_SIZE);
mol.EndModify();
mol.SetTitle(title);
return(true);
}
bool MOL2Format::ReadMolecule(OBBase* pOb, OBConversion* pConv)
{
OBMol* pmol = pOb->CastAndClear<OBMol>();
if(pmol==NULL)
return false;
//Define some references so we can use the old parameter names
istream &ifs = *pConv->GetInStream();
OBMol &mol = *pmol;
//Old code follows...
bool foundAtomLine = false;
char buffer[BUFF_SIZE];
char *comment = NULL;
string str,str1;
vector<string> vstr;
int len;
mol.BeginModify();
for (;;)
{
if (!ifs.getline(buffer,BUFF_SIZE))
return(false);
if (EQn(buffer,"@<TRIPOS>MOLECULE",17))
break;
}
int lcount;
int natoms,nbonds;
for (lcount=0;;lcount++)
{
if (!ifs.getline(buffer,BUFF_SIZE))
return(false);
if (EQn(buffer,"@<TRIPOS>ATOM",13))
{
foundAtomLine = true;
break;
}
if (lcount == 0)
{
tokenize(vstr,buffer);
if (!vstr.empty())
mol.SetTitle(buffer);
}
else if (lcount == 1)
sscanf(buffer,"%d%d",&natoms,&nbonds);
else if (lcount == 4) //energy
{
tokenize(vstr,buffer);
if (!vstr.empty() && vstr.size() == 3)
if (vstr[0] == "Energy")
mol.SetEnergy(atof(vstr[2].c_str()));
}
else if (lcount == 5) //comment
{
if ( buffer[0] )
{
len = (int) strlen(buffer)+1;
// TODO allow better multi-line comments
// which don't allow ill-formed data to consume memory
// Thanks to Andrew Dalke for the pointer
if (comment != NULL)
delete [] comment;
comment = new char [len];
memcpy(comment,buffer,len);
}
}
}
if (!foundAtomLine)
{
mol.EndModify();
mol.Clear();
obErrorLog.ThrowError(__FUNCTION__, "Unable to read Mol2 format file. No atoms found.", obWarning);
return(false);
}
mol.ReserveAtoms(natoms);
int i;
vector3 v;
OBAtom atom;
bool hasPartialCharges=false;
double x,y,z,pcharge;
char temp_type[BUFF_SIZE], resname[BUFF_SIZE], atmid[BUFF_SIZE];
int elemno, resnum = -1;
ttab.SetFromType("SYB");
for (i = 0;i < natoms;i++)
{
if (!ifs.getline(buffer,BUFF_SIZE))
return(false);
sscanf(buffer," %*s %1024s %lf %lf %lf %1024s %d %1024s %lf",
atmid, &x,&y,&z, temp_type, &resnum, resname, &pcharge);
atom.SetVector(x, y, z);
// Handle "CL" and "BR" and other mis-typed atoms
str = temp_type;
if (strncmp(temp_type, "CL", 2) == 0) {
str = "Cl";
} else if (strncmp(temp_type,"BR",2) == 0) {
str = "Br";
} else if (strncmp(temp_type,"S.o2", 4) == 02) {
str = "S.O2";
} else if (strncmp(temp_type,"S.o", 3) == 0) {
str = "S.O";
} else if (strncmp(temp_type,"SI", 2) == 0) {
str = "Si";
// The following cases are entries which are not in openbabel/data/types.txt
// and should probably be added there
} else if (strncmp(temp_type,"S.1", 3) == 0) {
str = "S.2"; // no idea what the best type might be here
} else if (strncmp(temp_type,"P.", 2) == 0) {
str = "P.3";
} else if (strncasecmp(temp_type,"Ti.", 3) == 0) { // e.g. Ti.th
str = "Ti";
} else if (strncasecmp(temp_type,"Ru.", 3) == 0) { // e.g. Ru.oh
str = "Ru";
}
ttab.SetToType("ATN");
ttab.Translate(str1,str);
elemno = atoi(str1.c_str());
ttab.SetToType("IDX");
// We might have missed some SI or FE type things above, so here's
// another check
if( !elemno && isupper(temp_type[1]) )
{
temp_type[1] = (char)tolower(temp_type[1]);
str = temp_type;
ttab.Translate(str1,str);
elemno = atoi(str1.c_str());
}
// One last check if there isn't a period in the type,
// it's a malformed atom type, but it may be the element symbol
// GaussView does this (PR#1739905)
if ( !elemno ) {
obErrorLog.ThrowError(__FUNCTION__, "This Mol2 file is non-standard. Cannot interpret atom types correctly, instead attempting to interpret as elements instead.", obWarning);
string::size_type dotPos = str.find('.');
if (dotPos == string::npos) {
elemno = etab.GetAtomicNum(str.c_str());
}
}
atom.SetAtomicNum(elemno);
ttab.SetToType("INT");
ttab.Translate(str1,str);
atom.SetType(str1);
atom.SetPartialCharge(pcharge);
if (!mol.AddAtom(atom))
return(false);
if (!IsNearZero(pcharge))
hasPartialCharges = true;
// Add residue information if it exists
if (resnum != -1 && resnum != 0 &&
strlen(resname) != 0 && strncmp(resname,"<1>", 3) != 0)
{
OBResidue *res = (mol.NumResidues() > 0) ?
mol.GetResidue(mol.NumResidues()-1) : NULL;
if (res == NULL || res->GetName() != resname ||
static_cast<int>(res->GetNum()) != resnum)
{
vector<OBResidue*>::iterator ri;
for (res = mol.BeginResidue(ri) ; res ; res = mol.NextResidue(ri))
if (res->GetName() == resname &&
static_cast<int>(res->GetNum())
== resnum)
break;
if (res == NULL)
{
res = mol.NewResidue();
res->SetName(resname);
res->SetNum(resnum);
}
}
OBAtom *atomPtr = mol.GetAtom(mol.NumAtoms());
res->AddAtom(atomPtr);
res->SetAtomID(atomPtr, atmid);
} // end adding residue info
}
for (;;)
{
if (!ifs.getline(buffer,BUFF_SIZE))
return(false);
str = buffer;
if (!strncmp(buffer,"@<TRIPOS>BOND",13))
break;
}
int start,end,order;
for (i = 0; i < nbonds; i++)
{
if (!ifs.getline(buffer,BUFF_SIZE))
return(false);
sscanf(buffer,"%*d %d %d %1024s",&start,&end,temp_type);
str = temp_type;
order = 1;
if (str == "ar" || str == "AR" || str == "Ar")
order = 5;
else if (str == "AM" || str == "am" || str == "Am")
order = 1;
else
order = atoi(str.c_str());
mol.AddBond(start,end,order);
}
// update neighbour bonds information for each atom.
vector<OBAtom*>::iterator apos;
vector<OBBond*>::iterator bpos;
OBAtom* patom;
OBBond* pbond;
for (patom = mol.BeginAtom(apos); patom; patom = mol.NextAtom(apos))
{
patom->ClearBond();
for (pbond = mol.BeginBond(bpos); pbond; pbond = mol.NextBond(bpos))
{
if (patom == pbond->GetBeginAtom() || patom == pbond->GetEndAtom())
{
patom->AddBond(pbond);
}
}
}
// Suggestion by Liu Zhiguo 2008-01-26
// Mol2 files define atom types -- there is no need to re-perceive
mol.SetAtomTypesPerceived();
mol.EndModify();
//must add generic data after end modify - otherwise it will be blown away
if (comment)
{
OBCommentData *cd = new OBCommentData;
cd->SetData(comment);
cd->SetOrigin(fileformatInput);
mol.SetData(cd);
delete [] comment;
comment = NULL;
}
if (hasPartialCharges)
mol.SetPartialChargesPerceived();
// continue untill EOF or untill next molecule record
streampos pos;
for(;;)
{
pos = ifs.tellg();
if (!ifs.getline(buffer,BUFF_SIZE))
break;
if (EQn(buffer,"@<TRIPOS>MOLECULE",17))
break;
}
ifs.seekg(pos); // go back to the end of the molecule
return(true);
}
bool HINFormat::ReadMolecule(OBBase* pOb, OBConversion* pConv)
{
OBMol* pmol = pOb->CastAndClear<OBMol>();
if(pmol==NULL)
return false;
//Define some references so we can use the old parameter names
istream &ifs = *pConv->GetInStream();
OBMol &mol = *pmol;
const char* title = pConv->GetTitle();
// Right now only read in the first molecule
int i;
int max, bo;
char buffer[BUFF_SIZE];
string str,str1;
double x,y,z;
OBAtom *atom;
vector<string> vs;
ifs.getline(buffer, BUFF_SIZE);
while (ifs.good() && (strstr(buffer,"mol") == NULL || buffer[0]==';') ) //The "mol" in comment line should be ignored.
{
ifs.getline(buffer, BUFF_SIZE);
if (ifs.peek() == EOF || !ifs.good())
return false;
}
ifs.getline(buffer, BUFF_SIZE);
if (!ifs.good())
return false; // ended early
mol.BeginModify();
while (ifs.good() && strstr(buffer,"endmol") == NULL)
{
if(buffer[0]==';'){
ifs.getline(buffer, BUFF_SIZE);
continue; //The comment Line in HIN should be ignored.
}
tokenize(vs,buffer); // Don't really know how long it'll be
if (vs.size() < 11)
{
ifs.getline(buffer, BUFF_SIZE);
continue;
}
atom = mol.NewAtom();
atom->SetAtomicNum(etab.GetAtomicNum(vs[3].c_str()));
atom->SetPartialCharge(atof(vs[6].c_str()));
x = atof((char*)vs[7].c_str());
y = atof((char*)vs[8].c_str());
z = atof((char*)vs[9].c_str());
atom->SetVector(x,y,z);
max = 11 + 2 * atoi((char *)vs[10].c_str());
for (i = 11; i < max; i+=2)
{
switch(((char*)vs[i+1].c_str())[0]) // First char in next token
{
case 's':
bo = 1;
break;
case 'd':
bo = 2;
break;
case 't':
bo = 3;
break;
case 'a':
bo = 5;
break;
default :
bo = 1;
break;
}
mol.AddBond(mol.NumAtoms(), atoi((char *)vs[i].c_str()), bo);
}
ifs.getline(buffer, BUFF_SIZE);
}
// clean out remaining blank lines
while(ifs.peek() != EOF && ifs.good() &&
(ifs.peek() == '\n' || ifs.peek() == '\r'))
ifs.getline(buffer,BUFF_SIZE);
mol.EndModify();
mol.SetTitle(title);
mol.SetPartialChargesPerceived();
return(true);
}