src/openbabel/xyzformat.cpp

/**********************************************************************
Copyright (C) 2000 by OpenEye Scientific Software, Inc.
Some portions Copyright (C) 2001-2005 by Geoffrey R. Hutchison
Some portions Copyright (C) 2004 by Chris Morley
 
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation version 2 of the License.
 
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.
***********************************************************************/

#include "mol.hpp"
#include "obconversion.hpp"
#include "obmolecformat.hpp"

#ifdef HAVE_SSTREAM
#include <sstream>
#else
#include <strstream>
#endif

using namespace std;
namespace OpenBabel
{

class XYZFormat : public OBMoleculeFormat
{
public:
  //Register this format type ID
  XYZFormat()
  {
    OBConversion::RegisterFormat("xyz", this, "chemical/x-xyz");
  }

  virtual const char* Description() //required
  {
    return
      "XYZ cartesian coordinates format\n \
       Read Options e.g. -as\n\
        s  Output single bonds only\n\
        b  Disable bonding entirely\n\n";
  };

  virtual const char* SpecificationURL()
  {return "http://openbabel.sourceforge.net/formats/xyz.shtml";}; //optional

  virtual const char* GetMIMEType() 
  { return "chemical/x-xyz"; };

  //*** This section identical for most OBMol conversions ***
  ////////////////////////////////////////////////////
  /// The "API" interface functions
  virtual bool ReadMolecule(OBBase* pOb, OBConversion* pConv);
  virtual bool WriteMolecule(OBBase* pOb, OBConversion* pConv);
};
//***

//Make an instance of the format class
XYZFormat theXYZFormat;

/////////////////////////////////////////////////////////////////
bool XYZFormat::ReadMolecule(OBBase* pOb, OBConversion* pConv)
{
    OBMol* pmol = dynamic_cast<OBMol*>(pOb);
    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];

#ifdef HAVE_SSTREAM
    stringstream errorMsg;
#else
    strstream errorMsg;
#endif

    unsigned int natoms;        // [ejk] assumed natoms could not be -ve

    if (!ifs.getline(buffer,BUFF_SIZE))
      {
        obErrorLog.ThrowError(__FUNCTION__,
                              "Problems reading an XYZ file: Cannot read the first line.", obWarning);
        return(false);
      }

    if (sscanf(buffer, "%d", &natoms) == 0 || !natoms)
    {
      obErrorLog.ThrowError(__FUNCTION__,
                            "Problems reading an XYZ file: The first line must contain the number of atoms.", obWarning);
      return(false);
    }

    mol.ReserveAtoms(natoms);

    // The next line contains a title string for the molecule. Use this
    // as the title for the molecule if the line is not
    // empty. Otherwise, use the title given by the calling function.
    if (!ifs.getline(buffer,BUFF_SIZE))
    {
      obErrorLog.ThrowError(__FUNCTION__,
                            "Problems reading an XYZ file: Could not read the second line (title/comments).", obWarning);
        return(false);
    }
    if (strlen(buffer) != 0)
        mol.SetTitle(buffer);
    else
        mol.SetTitle(title);

    mol.BeginModify();

    // The next lines contain four items each, separated by white
    // spaces: the atom type, and the coordinates of the atom
    vector<string> vs;
    for (unsigned int i = 1; i <= natoms; i ++)
    {
        if (!ifs.getline(buffer,BUFF_SIZE))
        {
          errorMsg << "Problems reading an XYZ file: "
                   << "Could not read line #" << i+2 << ", file error." << endl
                   << " According to line one, there should be " << natoms 
                   << " atoms, and therefore " << natoms+2 << " lines in the file.";

          obErrorLog.ThrowError(__FUNCTION__, errorMsg.str() , obWarning);
          return(false);
        }
        tokenize(vs,buffer);
        if (vs.size() != 4)
        {
          errorMsg << "Problems reading an XYZ file: "
                   << "Could not read line #" << i+2 << "." << endl
                   << "OpenBabel found the line '" << buffer << "'" << endl
                   << "According to the specifications, this line should contain exactly 4 entries, separated by white space." << endl
                   << "However, OpenBabel found " << vs.size() << " items.";

          obErrorLog.ThrowError(__FUNCTION__, errorMsg.str() , obWarning);
          return(false);
        }

        // Atom Type: get the atomic number from the element table, using
        // the first entry in the currently read line. If the entry makes
        // sense, set the atomic number and leave the atomic type open
        // (the type is then later faulted in when atom->GetType() is
        // called). If the entry does not make sense to use, set the atom
        // type manually, assuming that the author of the xyz-file had
        // something "special" in mind.
        OBAtom *atom  = mol.NewAtom();
        int atomicNum = etab.GetAtomicNum(vs[0].c_str());
        atom->SetAtomicNum(atomicNum); //set atomic number, or '0' if the atom type is not recognized
        if (atomicNum == 0)
            atom->SetType(vs[0]);

        // Read the atom coordinates
        char *endptr;
        double x = strtod((char*)vs[1].c_str(),&endptr);
        if (endptr == (char*)vs[1].c_str())
        {
          errorMsg << "Problems reading an XYZ file: "
                   << "Could not read line #" << i+2 << "." << endl
                   << "OpenBabel found the line '" << buffer << "'" << endl
                   << "According to the specifications, this line should contain exactly 4 entries, separated by white space." << endl
                   << "OpenBabel could not interpret item #1 as a number.";

          obErrorLog.ThrowError(__FUNCTION__, errorMsg.str() , obWarning);
          return(false);
        }
        double y = strtod((char*)vs[2].c_str(),&endptr);
        if (endptr == (char*)vs[2].c_str())
        {
          errorMsg << "Problems reading an XYZ file: "
                   << "Could not read line #" << i+2 << "." << endl
                   << "OpenBabel found the line '" << buffer << "'" << endl
                   << "According to the specifications, this line should contain exactly 4 entries, separated by white space." << endl
                   << "OpenBabel could not interpret item #2 as a number.";

          obErrorLog.ThrowError(__FUNCTION__, errorMsg.str() , obWarning);
          return(false);
        }
        double z = strtod((char*)vs[3].c_str(),&endptr);
        if (endptr == (char*)vs[3].c_str())
        {
          errorMsg << "Problems reading an XYZ file: "
                   << "Could not read line #" << i+2 << "." << endl
                   << "OpenBabel found the line '" << buffer << "'" << endl
                   << "According to the specifications, this line should contain exactly 4 entries, separated by white space." << endl
                   << "OpenBabel could not interpret item #3 as a number.";

          obErrorLog.ThrowError(__FUNCTION__, errorMsg.str() , obWarning);
          return(false);
        }
        atom->SetVector(x,y,z); //set coordinates
    }

    // 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();

    mol.EndModify();

    return(true);
}

////////////////////////////////////////////////////////////////

bool XYZFormat::WriteMolecule(OBBase* pOb, OBConversion* pConv)
{
    OBMol* pmol = dynamic_cast<OBMol*>(pOb);
    if(pmol==NULL)
        return false;

    //Define some references so we can use the old parameter names
    ostream &ofs = *pConv->GetOutStream();
    OBMol &mol = *pmol;

    unsigned int i;
    char buffer[BUFF_SIZE];

    sprintf(buffer,"%d", mol.NumAtoms());
    ofs << buffer << endl;
    sprintf(buffer,"%s\tEnergy: %15.7f", mol.GetTitle(), mol.GetEnergy());
    ofs << buffer << endl;

    OBAtom *atom;
    string str,str1;
    for(i = 1;i <= mol.NumAtoms(); i++)
    {
        atom = mol.GetAtom(i);
        sprintf(buffer,"%3s%15.5f%15.5f%15.5f",
                etab.GetSymbol(atom->GetAtomicNum()),
                atom->GetX(),
                atom->GetY(),
                atom->GetZ());
        ofs << buffer << endl;
    }

    return(true);
}

} //namespace OpenBabel
Revision:	2440
Committed:	Wed Nov 16 19:42:11 2005 UTC (18 years, 9 months ago) by tim
File size:	8514 byte(s)
Log Message:	adding openbabel
#	User	Rev	Content
1	tim	2440	/**********************************************************************
2			Copyright (C) 2000 by OpenEye Scientific Software, Inc.
3			Some portions Copyright (C) 2001-2005 by Geoffrey R. Hutchison
4			Some portions Copyright (C) 2004 by Chris Morley
5
6			This program is free software; you can redistribute it and/or modify
7			it under the terms of the GNU General Public License as published by
8			the Free Software Foundation version 2 of the License.
9
10			This program is distributed in the hope that it will be useful,
11			but WITHOUT ANY WARRANTY; without even the implied warranty of
12			MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13			GNU General Public License for more details.
14			***********************************************************************/
15
16			#include "mol.hpp"
17			#include "obconversion.hpp"
18			#include "obmolecformat.hpp"
19
20			#ifdef HAVE_SSTREAM
21			#include <sstream>
22			#else
23			#include <strstream>
24			#endif
25
26			using namespace std;
27			namespace OpenBabel
28			{
29
30			class XYZFormat : public OBMoleculeFormat
31			{
32			public:
33			//Register this format type ID
34			XYZFormat()
35			{
36			OBConversion::RegisterFormat("xyz", this, "chemical/x-xyz");
37			}
38
39			virtual const char* Description() //required
40			{
41			return
42			"XYZ cartesian coordinates format\n \
43			Read Options e.g. -as\n\
44			s Output single bonds only\n\
45			b Disable bonding entirely\n\n";
46			};
47
48			virtual const char* SpecificationURL()
49			{return "http://openbabel.sourceforge.net/formats/xyz.shtml";}; //optional
50
51			virtual const char* GetMIMEType()
52			{ return "chemical/x-xyz"; };
53
54			//* This section identical for most OBMol conversions *
55			////////////////////////////////////////////////////
56			/// The "API" interface functions
57			virtual bool ReadMolecule(OBBase* pOb, OBConversion* pConv);
58			virtual bool WriteMolecule(OBBase* pOb, OBConversion* pConv);
59			};
60			//***
61
62			//Make an instance of the format class
63			XYZFormat theXYZFormat;
64
65			/////////////////////////////////////////////////////////////////
66			bool XYZFormat::ReadMolecule(OBBase* pOb, OBConversion* pConv)
67			{
68			OBMol* pmol = dynamic_cast<OBMol*>(pOb);
69			if(pmol==NULL)
70			return false;
71
72			//Define some references so we can use the old parameter names
73			istream &ifs = *pConv->GetInStream();
74			OBMol &mol = *pmol;
75			const char* title = pConv->GetTitle();
76			char buffer[BUFF_SIZE];
77
78			#ifdef HAVE_SSTREAM
79			stringstream errorMsg;
80			#else
81			strstream errorMsg;
82			#endif
83
84			unsigned int natoms; // [ejk] assumed natoms could not be -ve
85
86			if (!ifs.getline(buffer,BUFF_SIZE))
87			{
88			obErrorLog.ThrowError(__FUNCTION__,
89			"Problems reading an XYZ file: Cannot read the first line.", obWarning);
90			return(false);
91			}
92
93			if (sscanf(buffer, "%d", &natoms) == 0 \|\| !natoms)
94			{
95			obErrorLog.ThrowError(__FUNCTION__,
96			"Problems reading an XYZ file: The first line must contain the number of atoms.", obWarning);
97			return(false);
98			}
99
100			mol.ReserveAtoms(natoms);
101
102			// The next line contains a title string for the molecule. Use this
103			// as the title for the molecule if the line is not
104			// empty. Otherwise, use the title given by the calling function.
105			if (!ifs.getline(buffer,BUFF_SIZE))
106			{
107			obErrorLog.ThrowError(__FUNCTION__,
108			"Problems reading an XYZ file: Could not read the second line (title/comments).", obWarning);
109			return(false);
110			}
111			if (strlen(buffer) != 0)
112			mol.SetTitle(buffer);
113			else
114			mol.SetTitle(title);
115
116			mol.BeginModify();
117
118			// The next lines contain four items each, separated by white
119			// spaces: the atom type, and the coordinates of the atom
120			vector<string> vs;
121			for (unsigned int i = 1; i <= natoms; i ++)
122			{
123			if (!ifs.getline(buffer,BUFF_SIZE))
124			{
125			errorMsg << "Problems reading an XYZ file: "
126			<< "Could not read line #" << i+2 << ", file error." << endl
127			<< " According to line one, there should be " << natoms
128			<< " atoms, and therefore " << natoms+2 << " lines in the file.";
129
130			obErrorLog.ThrowError(__FUNCTION__, errorMsg.str() , obWarning);
131			return(false);
132			}
133			tokenize(vs,buffer);
134			if (vs.size() != 4)
135			{
136			errorMsg << "Problems reading an XYZ file: "
137			<< "Could not read line #" << i+2 << "." << endl
138			<< "OpenBabel found the line '" << buffer << "'" << endl
139			<< "According to the specifications, this line should contain exactly 4 entries, separated by white space." << endl
140			<< "However, OpenBabel found " << vs.size() << " items.";
141
142			obErrorLog.ThrowError(__FUNCTION__, errorMsg.str() , obWarning);
143			return(false);
144			}
145
146			// Atom Type: get the atomic number from the element table, using
147			// the first entry in the currently read line. If the entry makes
148			// sense, set the atomic number and leave the atomic type open
149			// (the type is then later faulted in when atom->GetType() is
150			// called). If the entry does not make sense to use, set the atom
151			// type manually, assuming that the author of the xyz-file had
152			// something "special" in mind.
153			OBAtom *atom = mol.NewAtom();
154			int atomicNum = etab.GetAtomicNum(vs[0].c_str());
155			atom->SetAtomicNum(atomicNum); //set atomic number, or '0' if the atom type is not recognized
156			if (atomicNum == 0)
157			atom->SetType(vs[0]);
158
159			// Read the atom coordinates
160			char *endptr;
161			double x = strtod((char*)vs[1].c_str(),&endptr);
162			if (endptr == (char*)vs[1].c_str())
163			{
164			errorMsg << "Problems reading an XYZ file: "
165			<< "Could not read line #" << i+2 << "." << endl
166			<< "OpenBabel found the line '" << buffer << "'" << endl
167			<< "According to the specifications, this line should contain exactly 4 entries, separated by white space." << endl
168			<< "OpenBabel could not interpret item #1 as a number.";
169
170			obErrorLog.ThrowError(__FUNCTION__, errorMsg.str() , obWarning);
171			return(false);
172			}
173			double y = strtod((char*)vs[2].c_str(),&endptr);
174			if (endptr == (char*)vs[2].c_str())
175			{
176			errorMsg << "Problems reading an XYZ file: "
177			<< "Could not read line #" << i+2 << "." << endl
178			<< "OpenBabel found the line '" << buffer << "'" << endl
179			<< "According to the specifications, this line should contain exactly 4 entries, separated by white space." << endl
180			<< "OpenBabel could not interpret item #2 as a number.";
181
182			obErrorLog.ThrowError(__FUNCTION__, errorMsg.str() , obWarning);
183			return(false);
184			}
185			double z = strtod((char*)vs[3].c_str(),&endptr);
186			if (endptr == (char*)vs[3].c_str())
187			{
188			errorMsg << "Problems reading an XYZ file: "
189			<< "Could not read line #" << i+2 << "." << endl
190			<< "OpenBabel found the line '" << buffer << "'" << endl
191			<< "According to the specifications, this line should contain exactly 4 entries, separated by white space." << endl
192			<< "OpenBabel could not interpret item #3 as a number.";
193
194			obErrorLog.ThrowError(__FUNCTION__, errorMsg.str() , obWarning);
195			return(false);
196			}
197			atom->SetVector(x,y,z); //set coordinates
198			}
199
200			// clean out any remaining blank lines
201			while(ifs.peek() != EOF && ifs.good() &&
202			(ifs.peek() == '\n' \|\| ifs.peek() == '\r'))
203			ifs.getline(buffer,BUFF_SIZE);
204
205			if (!pConv->IsOption("b",OBConversion::INOPTIONS))
206			mol.ConnectTheDots();
207			if (!pConv->IsOption("s",OBConversion::INOPTIONS) && !pConv->IsOption("b",OBConversion::INOPTIONS))
208			mol.PerceiveBondOrders();
209
210			mol.EndModify();
211
212			return(true);
213			}
214
215			////////////////////////////////////////////////////////////////
216
217			bool XYZFormat::WriteMolecule(OBBase* pOb, OBConversion* pConv)
218			{
219			OBMol* pmol = dynamic_cast<OBMol*>(pOb);
220			if(pmol==NULL)
221			return false;
222
223			//Define some references so we can use the old parameter names
224			ostream &ofs = *pConv->GetOutStream();
225			OBMol &mol = *pmol;
226
227			unsigned int i;
228			char buffer[BUFF_SIZE];
229
230			sprintf(buffer,"%d", mol.NumAtoms());
231			ofs << buffer << endl;
232			sprintf(buffer,"%s\tEnergy: %15.7f", mol.GetTitle(), mol.GetEnergy());
233			ofs << buffer << endl;
234
235			OBAtom *atom;
236			string str,str1;
237			for(i = 1;i <= mol.NumAtoms(); i++)
238			{
239			atom = mol.GetAtom(i);
240			sprintf(buffer,"%3s%15.5f%15.5f%15.5f",
241			etab.GetSymbol(atom->GetAtomicNum()),
242			atom->GetX(),
243			atom->GetY(),
244			atom->GetZ());
245			ofs << buffer << endl;
246			}
247
248			return(true);
249			}
250
251			} //namespace OpenBabel