bool OutputFormat::ReadMolecule(OBBase* pOb, OBConversion* pConv)
{
// so we want to read through the file until we can figure out
// what program actually created it
// if we get to the end, emit a warning
istream &ifs = *pConv->GetInStream();
char buffer[BUFF_SIZE];
OBFormat *pFormat = NULL;
std::string formatName;
// the detection strings are from the Chemical MIME project
// http://chemical-mime.sourceforge.net/chemical-mime-data.html
while (ifs.getline(buffer,BUFF_SIZE)) {
if ((strstr(buffer,"GAMESS execution script") != NULL) ||
(strstr(buffer,"PC GAMESS") != NULL) ||
(strstr(buffer,"GAMESS VERSION") != NULL)) {
// GAMESS output
formatName = "gamout";
break;
} else if (strstr(buffer,"=== G A M E S S - U K === ") != NULL) {
// GAMESS-UK output
formatName = "gukout";
break;
} else if (strstr(buffer,"Gaussian, Inc") != NULL) {
// Gaussian output
formatName = "g03";
break;
} else if (strstr(buffer,"GENERAL UTILITY LATTICE PROGRAM") != NULL) {
// GULP output -- not currently supported
break;
} else if (strstr(buffer,"MOPAC") != NULL) {
// MOPAC output
formatName = "mopout";
break;
} else if (strstr(buffer,"Program PWSCF") != NULL) {
// PWSCF
formatName = "pwscf";
break;
} else if (strstr(buffer,"Welcome to Q-Chem") != NULL) {
// Q-Chem output
formatName = "qcout";
break;
} else if (strstr(buffer,"Amsterdam Density Functional") != NULL) {
// ADF output
// Determine the kind of ADF output
while (ifs.getline(buffer, BUFF_SIZE)) {
if (strstr(buffer, "| A D F |") != NULL) {
formatName = "adfout";
break;
} else if (strstr(buffer, "| B A N D |") != NULL) {
formatName = "adfband";
break;
} else if (strstr(buffer, "| D F T B |") != NULL) {
formatName = "adfdftb";
break;
} else if (strstr(buffer, "DFTB Engine") != NULL) {
// "| D F T B |" is no longer printed in ADF 2018
// Hopefully, "DFTB Engine" will work fine...
formatName = "adfdftb";
break;
}
}
break;
} else if (strstr(buffer,"Northwest Computational Chemistry") != NULL) {
// NWChem output
formatName = "nwo";
break;
} else if (strstr(buffer,"MPQC: Massively Parallel Quantum Chemistry") != NULL) {
// MPQC output
formatName = "mpqc";
break;
} else if (strstr(buffer,"PROGRAM SYSTEM MOLPRO") != NULL) {
// MOLPRO output
formatName = "mpo";
break;
} else if ((strstr(buffer,"Schrodinger, Inc.") != NULL) &&
(strstr(buffer,"Jaguar") != NULL)) {
// Jaguar
formatName = "jout";
break;
} else if (strstr(buffer, "ABINIT") != NULL) {
// Abinit
formatName = "abinit";
break;
} else if (strstr(buffer, "ACES2") != NULL) {
// ACESII
formatName = "acesout";
break;
} else if (strstr(buffer, "CRYSTAL06") != NULL ||
strstr(buffer, "CRYSTAL09") != NULL) {
// CRYSTAL09
formatName = "c09out";
break;
} else if (strstr(buffer, "* O R C A *") != NULL) {
// ORCA
formatName = "orca";
break;
} else if (strstr(buffer, "WELCOME TO SIESTA") != NULL) {
// SIESTA
formatName = "siesta";
break;
}
}
// if we assigned something above, let's try to find it
if (formatName.length())
pFormat = pConv->FindFormat(formatName);
if (pFormat) {
ifs.seekg (0, ios::beg); // reset the stream to the beginning
ifs.clear();
bool success = pFormat->ReadMolecule(pOb, pConv);
// Tag the molecule with the format (e.g., if a program wants to know the kind of "out" or "log" file)
// We have to do this *after* ReadMolecule returns, or the data might be cleared
if (pOb) {
OBPairData *dp = new OBPairData;
dp->SetAttribute("File Format");
dp->SetValue(formatName);
dp->SetOrigin(fileformatInput);
pOb->SetData(dp);
}
return success;
}
obErrorLog.ThrowError(__FUNCTION__,
"Problems reading an output file: Could not determine the format of this file. Please report it to the openbabel-discuss @ lists.sourceforge.net mailing list.", obError);
return(false); // we couldn't figure out the format
}
bool CMLReactFormat::DoElement(const string& name)
{
double val;
if(name=="reaction")
{
_spmol.reset();
_preact->SetTitle(_pxmlConv->GetAttribute("id"));
}
else if(name=="molecule")
{
string reference = _pxmlConv->GetAttribute("ref");
if(!reference.empty())
{
_spmol = IMols[reference];
pmol = _spmol.get();
if(!pmol)
{
cerr << " Molecule reference \"" << reference <<"\" not found" << endl;
return false;
}
}
else
{
shared_ptr<OBMol> sp(new OBMol);
OBFormat* pCMLFormat = OBConversion::FindFormat("cml");
if(!pCMLFormat)
return false;
_pxmlConv->_SkipNextRead=true;
pCMLFormat->ReadMolecule(sp.get(), _pxmlConv);
//Store smart pointers to all molecules in map
_spmol=sp;
AddMolToList(_spmol,IMols);
}
}
else if(name=="rateParameters")
{
_pRD = new OBRateData; //to store rate constant data
_preact->SetData(_pRD);
string rt = _pxmlConv->GetAttribute("reactionType");
OBRateData::reaction_type enumrt=OBRateData::ARRHENIUS;
if(rt=="arrhenius") enumrt=OBRateData::ARRHENIUS;
else if(rt=="lindermann") enumrt=OBRateData::LINDERMANN;
else if(rt=="troe") enumrt=OBRateData::TROE;
else if(rt=="sri") enumrt=OBRateData::SRI;
else if(rt=="threeBody") enumrt=OBRateData::THREEBODY;
else
obErrorLog.ThrowError(__FUNCTION__, rt + " is not a known reactionType", obWarning);
_pRD->ReactionType = enumrt;
if(_pxmlConv->GetAttribute("reversible")=="true")
_preact->SetReversible();
}
else if(_pRD && name=="A")
{
if(_pxmlConv->GetContentDouble(val))
_pRD->SetRate(OBRateData::A, val);
}
else if(_pRD && name=="n")
{
if(_pxmlConv->GetContentDouble(val))
_pRD->SetRate(OBRateData::n, val);
}
else if(_pRD && name=="E")
{
if(_pxmlConv->GetContentDouble(val))
_pRD->SetRate(OBRateData::E, val);
}
else if(_pRD && name=="loA")
{
if(_pxmlConv->GetContentDouble(val))
_pRD->SetLoRate(OBRateData::A, val);
}
else if(_pRD && name=="lon")
{
if(_pxmlConv->GetContentDouble(val))
_pRD->SetLoRate(OBRateData::n, val);
}
else if(_pRD && name=="loE")
{
if(_pxmlConv->GetContentDouble(val))
_pRD->SetLoRate(OBRateData::E, val);
}
else if(_pRD && name=="troeParams")
{
string txt(_pxmlConv->GetContent());
if(!txt.empty())
{
stringstream ss(txt);
for(int i=0;i<4;++i)
{
ss >>val;
_pRD->SetTroeParams(i, val);
}
}
bool RXNFormat::ReadMolecule(OBBase* pOb, OBConversion* pConv)
{
OBMol* pmol = pOb->CastAndClear<OBMol>();
if (pmol == NULL)
return false;
OBFormat* pMolFormat = pConv->FindFormat("MOL");
if (pMolFormat==NULL)
return false;
istream &ifs = *pConv->GetInStream();
string ln;
// When MDLFormat reads the last product it may also read and discard
// the line with $RXN for the next reaction. But it then sets $RXNread option.
if(pConv->IsOption("$RXNread"))
pConv->RemoveOption("$RXNread", OBConversion::OUTOPTIONS);
else
{
if (!getline(ifs,ln))
return(false);
if(Trim(ln).find("$RXN")!=0)
return false; //Has to start with $RXN
}
if (!getline(ifs,ln))
return false; //reaction title
pmol->SetTitle(Trim(ln));
if (!getline(ifs,ln))
return false; //creator
if (!getline(ifs, ln))
return false; //comment
// Originally the comment was added to the reaction via:
// pmol->SetComment(Trim(ln));
if (!getline(ifs, ln))
return false; // num reactants, products, and optionally agents
unsigned int nReactants = 0, nProducts = 0, nAgents = 0;
bool ok = ParseComponent(ln.c_str() + 0, &nReactants);
if (!ok)
return false;
ok = ParseComponent(ln.c_str() + 3, &nProducts);
if (!ok)
return false;
if (ln[6] != '\0') { // optional agents
ok = ParseComponent(ln.c_str() + 6, &nAgents);
if (!ok)
return false;
}
if(nReactants + nProducts + nAgents)
{
//Read the first $MOL. The others are read at the end of the previous MOL
if(!getline(ifs, ln))
return false;
if(Trim(ln).find("$MOL")==string::npos)
return false;
}
OBReactionFacade rxnfacade(pmol);
// Note: If we supported it, we could read each of the rxn components directly
// into the returned OBMol instead of having to do a copy. Unfortunately,
// this isn't possible at the moment (MOL format will need some work first).
// Here is some example code to do it:
//
//unsigned int old_numatoms = 0;
//unsigned int compid = 1;
//for (int i = 0; i<nReactants; i++)
//{
// //Read a MOL file using the same OBConversion object but with a different format
// if (!pMolFormat->ReadMolecule(pmol, pConv))
// obErrorLog.ThrowError(__FUNCTION__, "Failed to read a reactant", obWarning);
// unsigned int numatoms = pmol->NumAtoms();
// for (unsigned int idx = old_numatoms + 1; idx <= numatoms; ++idx) {
// OBAtom* atom = pmol->GetAtom(idx);
// rxnfacade.SetRole(atom, REACTANT);
// rxnfacade.SetComponentId(atom, compid);
// }
// old_numatoms = numatoms;
// compid++;
//}
const char* type[3] = {"a reactant", "a product", "an agent"};
OBReactionRole role;
unsigned int num_components;
for(unsigned int N=0; N<3; N++) {
switch(N) {
case 0:
role = REACTANT;
num_components = nReactants;
break;
case 1:
role = PRODUCT;
num_components = nProducts;
break;
case 2:
role = AGENT;
num_components = nAgents;
break;
}
for (int i=0; i<num_components; i++)
{
//Read a MOL file using the same OBConversion object but with a different format
OBMol mol;
if (!pMolFormat->ReadMolecule(&mol, pConv)) {
std::string error = "Failed to read ";
error += type[N];
obErrorLog.ThrowError(__FUNCTION__, error, obWarning);
continue;
}
if (mol.NumAtoms() == 0) {
OBAtom* dummy = mol.NewAtom(); // Treat the empty OBMol as having a single dummy atom
OBPairData *pd = new OBPairData();
pd->SetAttribute("rxndummy");
pd->SetValue("");
pd->SetOrigin(fileformatInput);
dummy->SetData(pd);
}
rxnfacade.AddComponent(&mol, role);
}
}
pmol->SetIsReaction();
return true;
}
bool PQSFormat::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];
char coord_file[256];
char full_coord_path[256]="\0";
ifstream coordFileStream;
double bohr_to_angstrom=1.0;
unsigned int input_style, atom_count=0; //CM i removed
bool geom_found;
geom_found=false;
while (!geom_found && ifs.getline(buffer,BUFF_SIZE))
{
lowerit(buffer); //look for geom except in title or text
if (strstr(buffer,"geom")!=NULL &&
(strncmp(buffer,"text",4)!=0 && strncmp(buffer,"titl",4)!=0))
{
geom_found=true;
lowerit(buffer);
if (strstr(buffer,"bohr")!=NULL)
bohr_to_angstrom=0.529177249;
else
bohr_to_angstrom=1.0;
input_style=0;
if (strstr(buffer,"=tx90")!=NULL)
input_style=1;
if (strstr(buffer,"=tx92")!=NULL)
input_style=0;
if (strstr(buffer,"=pqs" )!=NULL)
input_style=0;
if (strstr(buffer,"file=")!=NULL)
{ //external geometry file
strncpy(coord_file,strstr(buffer,"file=")+5, sizeof(coord_file));
coord_file[sizeof(coord_file) - 1] = '\0';
if (strrchr(coord_file,' ')!=NULL)
*strrchr(coord_file,' ')='\0';
if (coord_file[0]!='/')
{
strncpy(full_coord_path,title, sizeof(full_coord_path));
full_coord_path[sizeof(full_coord_path)-1] = '\0';
if (strrchr(full_coord_path,'/')!=NULL)
*(strrchr(full_coord_path,'/')+1)='\0';
else
full_coord_path[0] = '\0';
}
strcat(full_coord_path,coord_file);
full_coord_path[sizeof(full_coord_path) - 1] = '\0';
stringstream errorMsg;
errorMsg <<"External geometry file referenced: "<< \
full_coord_path<<endl;
obErrorLog.ThrowError(__FUNCTION__, errorMsg.str(), obInfo);
coordFileStream.open(full_coord_path);
if (!coordFileStream)
{
obErrorLog.ThrowError(__FUNCTION__, "Cannot read external geometry file!", obError);
return(false);
// exit (-1);
}
else
{
ifs.seekg(0, ios::end); //move .inp file pointer to the end of file
//New framework mods
OBConversion coordconv(&coordFileStream);
OBFormat* pFormat;
if (strstr(buffer,"=car" )!=NULL)
pFormat =OBConversion::FindFormat("BIOSYM");
if (strstr(buffer,"=hin" )!=NULL)
pFormat = OBConversion::FindFormat("HIN");
if (strstr(buffer,"=pdb" )!=NULL)
pFormat = OBConversion::FindFormat("PDB");
if (strstr(buffer,"=mop" )!=NULL)
pFormat = OBConversion::FindFormat("MOPAC");
return pFormat->ReadMolecule(&mol,&coordconv);
/* if (strstr(buffer,"=car" )!=NULL)
return ReadBiosymCAR(coordFileStream, mol, title);
if (strstr(buffer,"=hin" )!=NULL)
return ReadHIN(coordFileStream, mol, title);
if (strstr(buffer,"=pdb" )!=NULL)
return ReadPDB(coordFileStream, mol, title);
if (strstr(buffer,"=mop" )!=NULL)
return ReadMOPAC(coordFileStream, mol, title);
*/
//end new framework mods
//probably pqs's own xyz format
atom_count=ReadPQS_geom(coordFileStream,mol,title,
input_style,bohr_to_angstrom);
}
}
}
}
if (geom_found)
{
if (atom_count==0) //read directly form .inp file
atom_count=ReadPQS_geom(ifs,mol,title,input_style,bohr_to_angstrom);
if (atom_count==0)
{ //try .coord file
strncpy(coord_file,title, sizeof(coord_file));
coord_file[sizeof(coord_file) - 1] = '\0';
if (strrchr(coord_file,'.')!=NULL)
*strrchr(coord_file,'.')='\0';
strcat(coord_file,".coord");
coordFileStream.open(coord_file);
if (!coordFileStream)
{
stringstream errorMsg;
errorMsg <<"ReadPQS: cannot read external "<<coord_file<<" file!"<<endl;
obErrorLog.ThrowError(__FUNCTION__, errorMsg.str(), obError);
return(false);
// exit (-1);
}
else
atom_count=ReadPQS_geom(coordFileStream,mol,title,0,
bohr_to_angstrom);
}
}
else
obErrorLog.ThrowError(__FUNCTION__, "Error reading PQS file. GEOM card not found!", obWarning);
ifs.seekg(0, ios::end); //move .inp file pointer to the end of file
if (atom_count>0)
return true;
else
return false;
}