re-ref started

git-svn-id: svn+ssh://gitlab/srv/svn-repos/pdb-redo/trunk@237 a1961a4f-ab94-4bcc-80e8-33b5a54de466

include/cif++/Cif++.h

@@ -252,10 +252,7 @@ namespace detail
 		template<typename T>
 		T as() const
 		{
-			T result = 0;
-			if (not empty())
-				result = boost::lexical_cast<T>(c_str());
-			return result;
+			return boost::lexical_cast<T>(c_str("0"));
 		}
 		
 		template<typename T>
@@ -285,6 +282,10 @@ namespace detail
 		
 		const char* c_str() const;
 		
+		// the following returns the defaultValue from either the parameter
+		// or, if specified, the value from _item_default.value in the dictionary
+		const char* c_str(const char* defaultValue) const;
+		
 		bool operator!=(const string& s) const		{ return s != c_str(); }
 		bool operator==(const string& s) const		{ return s == c_str(); }
 	};
@@ -293,14 +294,14 @@ namespace detail
 	inline
 	string ItemReference::as<string>() const
 	{
-		return string(c_str());
+		return string(c_str(""));
 	}
 	
 	template<>
 	inline
 	const char* ItemReference::as<const char*>() const
 	{
-		return c_str();
+		return c_str("");
 	}
 	
 	template<>
@@ -908,7 +909,14 @@ Condition any::operator==(const std::regex& rx) const
 
 class RowSet : public vector<Row>
 {
+	typedef vector<Row>	base_type;
+
   public:
+	RowSet(const RowSet& rhs);
+	RowSet(RowSet&& rhs);
+	RowSet& operator=(const RowSet& rhs);
+	RowSet& operator=(RowSet&& rhs);
+
 	RowSet(Category& cat);
 	
 	RowSet& orderBy(const string& Item)
@@ -917,7 +925,7 @@ class RowSet : public vector<Row>
 	RowSet& orderBy(std::initializer_list<string> Items);
 
   private:
-	Category&	mCat;
+	Category*	mCat;
 };
 
 // --------------------------------------------------------------------

include/cif++/CifValidator.h

@@ -58,8 +58,9 @@ struct ValidateItem
 {
 	std::string				mTag;
 	bool					mMandatory;
-	const ValidateType*	mType;
+	const ValidateType*		mType;
 	cif::iset				mEnums;
+	std::string				mDefault;
 	ValidateItem*			mParent = nullptr;
 	std::set<ValidateItem*>
 							mChildren;

include/cif++/Compound.h

@@ -104,7 +104,7 @@ class Compound
 	std::string id() const					{ return mId; }
 	std::string	name() const				{ return mName; }
 	std::string	type() const;
-//	std::string group() const				{ return mGroup; }
+	std::string group() const				{ return mGroup; }
 	std::vector<CompoundAtom> atoms() const	{ return mAtoms; }
 	std::vector<CompoundBond> bonds() const	{ return mBonds; }
 	

include/cif++/MapMaker.h

@@ -38,8 +38,7 @@ class MapMaker
 	void recalculateFromMTZ(const boost::filesystem::path& mtzFile,
 		const Structure& structure,
 		bool noBulk, AnisoScalingFlag anisoScaling,
-		float samplingRate = 4.5,
-		bool electronScattering = false,
+		float samplingRate = 4.5, bool electronScattering = false,
 		std::initializer_list<std::string> foLabels = { "FP", "SIGFP" },
 		std::initializer_list<std::string> freeLabels = { "FREE" });
 	void loadFromMapFiles(const boost::filesystem::path& fbMapFile,

include/cif++/Structure.h

@@ -167,6 +167,8 @@ class Residue
 	
 	// Is this residue a single entity?
 	bool isEntity() const;
+	
+	const Structure& structure() const		{ return *mStructure; }
 
   protected:
 
@@ -260,8 +262,8 @@ class Polymer
 	iterator begin();
 	iterator end();
 	
-	size_t size() const				{ return mPolySeq.size(); }
-	Monomer operator[](size_t index) const;
+	size_t size() const							{ return mPolySeq.size(); }
+	Monomer operator[](size_t index) const		{ return Monomer(const_cast<Polymer&>(*this), index); }
 
 	Structure* structure() const	{ return mStructure; }
 	

src/BondMap.cpp

@@ -102,13 +102,21 @@ BondMap::BondMap(const Structure& p)
 				  "ptnr1_label_atom_id", "ptnr2_label_atom_id",
 				  "ptnr1_label_seq_id", "ptnr2_label_seq_id");
 		
-		auto a = db["atom_site"].find(cif::Key("label_asym_id") == asym1 and cif::Key("label_seq_id") == seqId1 and cif::Key("label_atom_id") == atomId1);
+		auto a = 
+			l["ptnr1_label_seq_id"].empty() ?
+				db["atom_site"].find(cif::Key("label_asym_id") == asym1 and cif::Key("label_atom_id") == atomId1) :
+				db["atom_site"].find(cif::Key("label_asym_id") == asym1 and cif::Key("label_seq_id") == seqId1 and cif::Key("label_atom_id") == atomId1);
+		
 		if (a.size() != 1)
-			cerr << "Unexpected number (" << a.size() << ") of atoms for link with asym_id " << asym1 << " seq_id " << seqId1 << " atom_id  " << atomId1 << endl;
+			cerr << "Unexpected number (" << a.size() << ") of atoms for link with asym_id " << asym1 << " seq_id " << seqId1 << " atom_id " << atomId1 << endl;
 		
-		auto b = db["atom_site"].find(cif::Key("label_asym_id") == asym2 and cif::Key("label_seq_id") == seqId2 and cif::Key("label_atom_id") == atomId2);
+		auto b =
+			l["ptnr2_label_seq_id"].empty() ?
+				db["atom_site"].find(cif::Key("label_asym_id") == asym2 and cif::Key("label_atom_id") == atomId2) :
+				db["atom_site"].find(cif::Key("label_asym_id") == asym2 and cif::Key("label_seq_id") == seqId2 and cif::Key("label_atom_id") == atomId2);
+
 		if (b.size() != 1)
-			cerr << "Unexpected number (" << b.size() << ") of atoms for link with asym_id " << asym2 << " seq_id " << seqId2 << " atom_id  " << atomId2 << endl;
+			cerr << "Unexpected number (" << b.size() << ") of atoms for link with asym_id " << asym2 << " seq_id " << seqId2 << " atom_id " << atomId2 << endl;
 		
 		if (not (a.empty() or b.empty()))
 			bindAtoms(a.front()["id"].as<string>(), b.front()["id"].as<string>());

src/Cif++.cpp

@@ -90,7 +90,7 @@ void ItemValue::operator delete(void* p)
 struct ItemColumn
 {
 	string					mName;		// store lower-case, for optimization
-	const ValidateItem*	mValidator;
+	const ValidateItem*		mValidator;
 };
 
 // itemRow contains the actual values for a Row in a Category
@@ -183,7 +183,6 @@ void ItemRow::drop(uint32 columnIx)
 #if DEBUG
 	for (auto iv = mValues; iv != nullptr; iv = iv->mNext)
 		assert(iv != iv->mNext and (iv->mNext == nullptr or iv != iv->mNext->mNext));
-	
 #endif
 }
 
@@ -215,8 +214,7 @@ ItemReference& ItemReference::operator=(const string& value)
 	return *this;
 }
 
-const char*
-ItemReference::c_str() const
+const char* ItemReference::c_str() const
 {
 	const char* result = kEmptyResult;
 	
@@ -241,6 +239,39 @@ ItemReference::c_str() const
 	return result;
 }
 
+const char* ItemReference::c_str(const char* defaultValue) const
+{
+	const char* result = defaultValue;
+	
+	if (mRow != nullptr and mRow->mCategory != nullptr)
+	{
+//		assert(mRow->mCategory);
+		
+		auto cix = mRow->mCategory->getColumnIndex(mName);
+		
+		if (cix < mRow->mCategory->mColumns.size())
+		{
+			auto iv = mRow->mCategory->mColumns[cix].mValidator;
+			if (iv != nullptr and not iv->mDefault.empty())
+				result = iv->mDefault.c_str();
+			
+			for (auto iv = mRow->mValues; iv != nullptr; iv = iv->mNext)
+			{
+				if (iv->mColumnIndex == cix)
+				{
+					// only really non-empty values
+					if (iv->mText[0] != 0 and ((iv->mText[0] != '.' and iv->mText[0] != '?') or iv->mText[1] != 0))
+						result = iv->mText;
+	
+					break;
+				}
+			}
+		}
+	}
+	
+	return result;
+}
+
 bool ItemReference::empty() const
 {
 	return c_str() == kEmptyResult;
@@ -991,14 +1022,48 @@ void CatIndex::validate(entry* h, bool isParentRed, uint32 blackDepth, uint32& m
 
 // --------------------------------------------------------------------
 
+RowSet::RowSet(const RowSet& rhs)
+	: base_type(rhs)
+	, mCat(rhs.mCat)
+{
+}
+
+RowSet::RowSet(RowSet&& rhs)
+	: base_type(move(rhs))
+	, mCat(rhs.mCat)
+{
+}
+
+RowSet& RowSet::operator=(const RowSet& rhs)
+{
+	if (this != &rhs)
+	{
+		base_type::operator=(rhs);
+		mCat = rhs.mCat;
+	}
+	
+	return *this;
+}
+
+RowSet& RowSet::operator=(RowSet&& rhs)
+{
+	if (this != &rhs)
+	{
+		base_type::operator=(move(rhs));
+		mCat = rhs.mCat;
+	}
+	
+	return *this;
+}
+
 RowSet::RowSet(Category& cat)
-	: mCat(cat)
+	: mCat(&cat)
 {
 }
 
 RowSet& RowSet::orderBy(initializer_list<string> items)
 {
-	RowComparator c(&mCat, items.begin(), items.end());
+	RowComparator c(mCat, items.begin(), items.end());
 	
 	stable_sort(begin(), end(), c);
 	

src/CifParser.cpp

@@ -784,6 +784,8 @@ void DictParser::parseSaveFrame()
 		for (auto e: dict["item_enumeration"])
 			ess.insert(e["value"].as<string>());
 		
+		string defaultValue = dict.firstItem("_item_default.value");
+		
 		// collect the dict from our dataBlock and construct validators
 		for (auto i: dict["item"])
 		{
@@ -806,7 +808,7 @@ void DictParser::parseSaveFrame()
 			
 			auto vi = find(ivs.begin(), ivs.end(), ValidateItem{itemName});
 			if (vi == ivs.end())
-				ivs.push_back(ValidateItem{itemName, iequals(mandatory, "yes"), tv, ess});
+				ivs.push_back(ValidateItem{itemName, iequals(mandatory, "yes"), tv, ess, defaultValue});
 			else
 			{
 				// need to update the itemValidator?

src/Compound.cpp

@@ -420,8 +420,8 @@ string Compound::type() const
 		result = "DNA linking";
 	else if (cif::iequals(mGroup, "RNA"))
 		result = "RNA linking";
-	else
-		result = mGroup;
+//	else
+//		result = mGroup;
 	
 	return result;
 }

src/MapMaker.cpp

@@ -108,7 +108,7 @@ void MapMaker<FTYPE>::loadFromMTZ(const fs::path& mtzFile, float samplingRate,
 template<typename FTYPE>
 void MapMaker<FTYPE>::fixMTZ(FPdata& fb, FPdata& fd, FOdata& fo, FPdata& fc)
 {
-#warning("WARNING: Need the check first to see if fix is necessary!")
+#warning("WARNING: Need to check first to see if fix is necessary!")
 
 	using clipper::HKL_class;
 	using clipper::data32::F_phi;
@@ -195,7 +195,6 @@ void MapMaker<FTYPE>::recalculateFromMTZ(const fs::path& mtzFile,
 
 	HKL_data<F_phi> fc(hklInfo, mCell);
 
-
 	if (not electronScattering)
 	{
 		auto& exptl = structure.getFile().data()["exptl"];

src/PDB2Cif.cpp

@@ -106,7 +106,10 @@ const set<string> kSupportedRecords{
 	"SSBOND", "LINK  ", "CISPEP", "SITE  ", "CRYST1", "ORIGX1", "SCALE1",
 	"MTRIX1", "ORIGX2", "SCALE2", "MTRIX2", "ORIGX3", "SCALE3", "MTRIX3",
 	"MODEL ", "ATOM  ", "ANISOU", "TER   ", "HETATM", "ENDMDL", "CONECT",
-	"MASTER", "END   "
+	"MASTER", "END   ",
+	
+	// bah...
+	"LINKR "
 };
 
 bool isWater(const string& resname)
@@ -4549,8 +4552,11 @@ void PDBFileParser::ParseConnectivtyAnnotation()
 			continue;
 		}
 		
-		if (mRec->is("LINK  "))
+		if (mRec->is("LINK  ") or mRec->is("LINKR "))
 		{
+			if (VERBOSE and mRec->is("LINKR "))
+				cerr << "Accepting non-standard LINKR record, but ignoring extra information" << endl;
+			
 											//	 1 -  6         Record name    "LINK  "
 			string name1 = vS(13, 16);		//	13 - 16         Atom           name1           Atom name.
 											//	17              Character      altLoc1         Alternate location indicator.
@@ -4606,6 +4612,18 @@ void PDBFileParser::ParseConnectivtyAnnotation()
 				continue;
 			}
 			
+			string distance = vS(74, 78);
+			try
+			{
+				stod(distance);
+			}
+			catch (const invalid_argument&)
+			{
+				if (VERBOSE)
+					cerr << "Distance value '" << distance << "' is not a valid float" << endl;
+				distance.clear();
+			}
+			
 			getCategory("struct_conn")->emplace({
 				{ "id", type + to_string(linkNr) },
 				{ "conn_type_id", type },
@@ -4635,7 +4653,7 @@ void PDBFileParser::ParseConnectivtyAnnotation()
 	
 				{ "ptnr2_symmetry", pdb2cifSymmetry(vS(67, 72)) },
 				
-				{ "pdbx_dist_value", vS(74, 78) }
+				{ "pdbx_dist_value", distance }
 			});
 			
 			continue;
@@ -5412,7 +5430,6 @@ int PDBFileParser::PDBChain::AlignResToSeqRes()
 	if (VERBOSE > 1)
 		printAlignment();
 	
-	
 	try
 	{
 		while (x >= 0 and y >= 0)
@@ -5462,9 +5479,6 @@ int PDBFileParser::PDBChain::AlignResToSeqRes()
 		throw;
 	}
 
-	if (VERBOSE > 1)
-		printAlignment();
-	
 	// assign numbers to the residues that don't have them yet
 	stack<int> unnumbered;
 	for (x = 0; x < dimX; ++x)