/*
 *                    BioJava development code
 *
 * This code may be freely distributed and modified under the
 * terms of the GNU Lesser General Public Licence.  This should
 * be distributed with the code.  If you do not have a copy,
 * see:
 *
 *      http://www.gnu.org/copyleft/lesser.html
 *
 * Copyright for this code is held jointly by the individual
 * authors.  These should be listed in @author doc comments.
 *
 * For more information on the BioJava project and its aims,
 * or to join the biojava-l mailing list, visit the home page
 * at:
 *
 *      http://www.biojava.org/
 *
 */
package org.biojava.nbio.structure.align.util;

import java.io.IOException;
import java.util.ArrayList;
import java.util.Collection;
import java.util.Collections;
import java.util.List;
import java.util.TreeSet;

import org.biojava.nbio.core.util.InputStreamProvider;
import org.biojava.nbio.structure.*;
import org.biojava.nbio.structure.align.client.StructureName;
import org.biojava.nbio.structure.cath.CathDatabase;
import org.biojava.nbio.structure.cath.CathDomain;
import org.biojava.nbio.structure.cath.CathFactory;
import org.biojava.nbio.structure.io.BcifFileReader;
import org.biojava.nbio.structure.io.CifFileReader;
import org.biojava.nbio.structure.io.FileParsingParameters;
import org.biojava.nbio.structure.io.LocalPDBDirectory.FetchBehavior;
import org.biojava.nbio.structure.io.LocalPDBDirectory.ObsoleteBehavior;
import org.biojava.nbio.structure.io.PDBFileReader;
import org.biojava.nbio.core.util.FileDownloadUtils;
import org.biojava.nbio.structure.io.StructureFiletype;
import org.biojava.nbio.structure.quaternary.BiologicalAssemblyBuilder;
import org.biojava.nbio.structure.quaternary.BiologicalAssemblyTransformation;
import org.biojava.nbio.structure.scop.ScopDatabase;
import org.biojava.nbio.structure.scop.ScopDescription;
import org.biojava.nbio.structure.scop.ScopDomain;
import org.biojava.nbio.structure.scop.ScopFactory;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

/**
 * A utility class that provides easy access to Structure objects. If you are running a script that is frequently
 * re-using the same PDB structures, the AtomCache keeps an in-memory cache of the files for quicker access. The cache
 * is a soft-cache, this means it won't cause out of memory exceptions, but garbage collects the data if the Java
 * virtual machine needs to free up space. The AtomCache is thread-safe.
 *
 * @author Andreas Prlic
 * @author Spencer Bliven
 * @author Peter Rose
 * @since 3.0
 */
public class AtomCache {
        private static final Logger logger = LoggerFactory.getLogger(AtomCache.class);

        /**
         * The default output bioassembly style: if true the bioassemblies are multimodel,
         * if false the bioassemblies are flat with renamed chains for symmetry-partners.
         */
        public static final boolean DEFAULT_BIOASSEMBLY_STYLE = false;

        public static final String BIOL_ASSEMBLY_IDENTIFIER = "BIO:";
        public static final String CHAIN_NR_SYMBOL = ":";
        public static final String CHAIN_SPLIT_SYMBOL = ".";
        public static final String UNDERSCORE = "_";

        private static final String FILE_SEPARATOR = System.getProperty("file.separator");

        protected FileParsingParameters params;
        private FetchBehavior fetchBehavior;
        private ObsoleteBehavior obsoleteBehavior;
        private String cachePath;

        // make sure IDs are loaded uniquely
        private final Collection<String> currentlyLoading = Collections.synchronizedCollection(new TreeSet<>());

        private String path;
        private StructureFiletype filetype = StructureFiletype.BCIF;

        /**
         * Default AtomCache constructor.
         *
         * Usually stores files in a temp directory, but this can be overriden by setting the PDB_DIR variable at runtime.
         *
         * @see UserConfiguration#UserConfiguration()
         */
        public AtomCache() {
                this(new UserConfiguration());
        }

        /**
         * Creates an instance of an AtomCache that is pointed to the a particular path in the file system. It will use the same value for pdbFilePath and cachePath.
         *
         * @param pdbFilePath
         *            a directory in the file system to use as a location to cache files.
         */
        public AtomCache(String pdbFilePath) {
                this(pdbFilePath,pdbFilePath);
        }

        /**
         * Creates an instance of an AtomCache that is pointed to the a particular path in the file system.
         *
         * @param pdbFilePath
         *            a directory in the file system to use as a location to cache files.
         * @param cachePath
         */
        public AtomCache(String pdbFilePath, String cachePath) {
                logger.debug("Initialising AtomCache with pdbFilePath={}, cachePath={}", pdbFilePath, cachePath);
                if (!pdbFilePath.endsWith(FILE_SEPARATOR)) {
                        pdbFilePath += FILE_SEPARATOR;
                }

                // we are caching the binary files that contain the PDBs gzipped
                // that is the most memory efficient way of caching...
                // set the input stream provider to caching mode
                System.setProperty(InputStreamProvider.CACHE_PROPERTY, "true");

                setPath(pdbFilePath);

                this.cachePath = cachePath;

                fetchBehavior = FetchBehavior.DEFAULT;
                obsoleteBehavior = ObsoleteBehavior.DEFAULT;

                currentlyLoading.clear();
                params = new FileParsingParameters();

                setFiletype(StructureFiletype.BCIF);
        }

        /**
         * Creates a new AtomCache object based on the provided UserConfiguration.
         *
         * @param config
         *            the UserConfiguration to use for this cache.
         */
        public AtomCache(UserConfiguration config) {
                this(config.getPdbFilePath(), config.getCacheFilePath());
                fetchBehavior = config.getFetchBehavior();
                obsoleteBehavior = config.getObsoleteBehavior();
                filetype = config.getStructureFiletype();
        }

        /**
         * Returns the CA atoms for the provided name. See {@link #getStructure(String)} for supported naming conventions.
         * <p>
         * This method only works with protein chains. Use {@link #getRepresentativeAtoms(String)}
         * for a more general solution.
         * @param name
         * @return an array of Atoms.
         * @throws IOException
         * @throws StructureException
         */
        public Atom[] getAtoms(String name) throws IOException, StructureException {
                return getAtoms(new StructureName(name));
        }

        public Atom[] getAtoms(StructureIdentifier name) throws IOException, StructureException {
                Atom[] atoms;

                // System.out.println("loading " + name);
                Structure s = getStructure(name);
                atoms = StructureTools.getAtomCAArray(s);

                /*
                 * synchronized (cache){ cache.put(name, atoms); }
                 */
                return atoms;
        }

        /**
         * Returns the representative atoms for the provided name.
         * See {@link #getStructure(String)} for supported naming conventions.
         *
         * @param name
         * @return an array of Atoms.
         * @throws IOException
         * @throws StructureException
         */
        public Atom[] getRepresentativeAtoms(String name) throws IOException, StructureException {
                return getRepresentativeAtoms(new StructureName(name));
        }

        public Atom[] getRepresentativeAtoms(StructureIdentifier name) throws IOException, StructureException {
                Atom[] atoms;

                Structure s = getStructure(name);
                atoms = StructureTools.getRepresentativeAtomArray(s);

                /*
                 * synchronized (cache){ cache.put(name, atoms); }
                 */
                return atoms;
        }

        /**
         * Returns the biological assembly for a given PDB ID and bioAssemblyId, by building the
         * assembly from the biounit annotations found in {@link Structure#getPDBHeader()}
         * <p>
         * Note, the number of available biological unit files
         * varies. Many entries don't have a biological assembly specified (e.g. NMR structures), many entries have only one
         * biological assembly (bioAssemblyId=1), and some structures have multiple biological assemblies.
         *
         * @param pdbId
         *            the PDB ID
         * @param bioAssemblyId
         *            the 1-based index of the biological assembly (0 gets the asymmetric unit)
         * @param multiModel if true the output Structure will be a multi-model one with one transformId per model,
         * if false the outputStructure will be as the original with added chains with renamed asymIds (in the form originalAsymId_transformId and originalAuthId_transformId).
         * @return a structure object
         * @throws IOException
         * @throws StructureException if biassemblyId &lt; 0 or other problems while loading structure
         * @since 3.2
         */
        public Structure getBiologicalAssembly(String pdbId, int bioAssemblyId, boolean multiModel)
                        throws StructureException, IOException {
                return getBiologicalAssembly(new PdbId(pdbId), bioAssemblyId, multiModel);
        }
        
        /**
         * Returns the biological assembly for a given PDB ID and bioAssemblyId, by building the
         * assembly from the biounit annotations found in {@link Structure#getPDBHeader()}
         * <p>
         * Note, the number of available biological unit files
         * varies. Many entries don't have a biological assembly specified (e.g. NMR structures), many entries have only one
         * biological assembly (bioAssemblyId=1), and some structures have multiple biological assemblies.
         *
         * @param pdbId
         *            the PDB ID
         * @param bioAssemblyId
         *            the 1-based index of the biological assembly (0 gets the asymmetric unit)
         * @param multiModel if true the output Structure will be a multi-model one with one transformId per model,
         * if false the outputStructure will be as the original with added chains with renamed asymIds (in the form originalAsymId_transformId and originalAuthId_transformId).
         * @return a structure object
         * @throws IOException
         * @throws StructureException if biassemblyId &lt; 0 or other problems while loading structure
         * @since 6.0.0
         */
        public Structure getBiologicalAssembly(PdbId pdbId, int bioAssemblyId, boolean multiModel)
                        throws StructureException, IOException {
                if (bioAssemblyId < 0) {
                        throw new StructureException("bioAssemblyID must be nonnegative: " + pdbId + " bioAssemblyId "
                                        + bioAssemblyId);
                }

                boolean prevIsParseBioAssembly = getFileParsingParams().isParseBioAssembly();

                if (!getFileParsingParams().isParseBioAssembly()) {
                        getFileParsingParams().setParseBioAssembly(true);
                }

                Structure asymUnit = getStructureForPdbId(pdbId);

                getFileParsingParams().setParseBioAssembly(prevIsParseBioAssembly);

                if (asymUnit.getPDBHeader() == null || asymUnit.getPDBHeader().getBioAssemblies() == null) {
                        logger.info("No bioassembly information found for {}, returning asymmetric unit as biological assembly", pdbId);
                        return asymUnit;
                }

                // 0 ... asym unit
                if (bioAssemblyId == 0) {
                        logger.info("Requested biological assembly 0 for PDB id {}, returning asymmetric unit", pdbId);
                        return asymUnit;
                }
                // does it exist?
                if (!asymUnit.getPDBHeader().getBioAssemblies().containsKey(bioAssemblyId)) {
                        throw new StructureException("No biological assembly available for biological assembly id " + bioAssemblyId + " of " + pdbId);
                }

                List<BiologicalAssemblyTransformation> transformations =
                                asymUnit.getPDBHeader().getBioAssemblies().get(bioAssemblyId).getTransforms();


                if (transformations == null || transformations.size() == 0) {
                        throw new StructureException("Could not load transformations to recreate biological assembly id " + bioAssemblyId + " of " + pdbId);
                }

                BiologicalAssemblyBuilder builder = new BiologicalAssemblyBuilder();

                // if we use mmcif or mmtf, then we need to pass useAsymIds=true
                boolean useAsymIds = false;
                if (filetype == StructureFiletype.CIF || filetype == StructureFiletype.BCIF || filetype == StructureFiletype.MMTF) {
                        useAsymIds = true;
                }
                return builder.rebuildQuaternaryStructure(asymUnit, transformations, useAsymIds, multiModel);
        }

        /**
         * Returns the default biological unit (bioassemblyId=1, known in PDB as pdb1.gz). If it is not available,
         * the asymmetric unit will be returned, e.g. for NMR structures.
         *
         * <p>Biological assemblies can also be accessed using
         * <code>getStructure("BIO:<i>[pdbId]</i>")</code>
         * @param pdbId the PDB id
         * @param multiModel if true the output Structure will be a multi-model one with one transformId per model,
         * if false the outputStructure will be as the original with added chains with renamed asymIds (in the form originalAsymId_transformId and originalAuthId_transformId).
         * @return a structure object
         * @throws IOException
         * @throws StructureException
         * @since 4.2
         */
        public Structure getBiologicalAssembly(String pdbId, boolean multiModel) throws StructureException, IOException {
                boolean prevIsParseBioAssembly = getFileParsingParams().isParseBioAssembly();

                if (!getFileParsingParams().isParseBioAssembly()) {
                        getFileParsingParams().setParseBioAssembly(true);
                }

                Structure asymUnit = getStructureForPdbId(pdbId);
                getFileParsingParams().setParseBioAssembly(prevIsParseBioAssembly);


                if (asymUnit.getPDBHeader() == null || asymUnit.getPDBHeader().getBioAssemblies() == null) {
                        logger.info("No bioassembly information found for {}, returning asymmetric unit as biological assembly", pdbId);
                        return asymUnit;
                }

                int bioAssemblyId = 1;

                // does it exist?
                if (!asymUnit.getPDBHeader().getBioAssemblies().containsKey(bioAssemblyId)) {
                        return asymUnit;
                }

                List<BiologicalAssemblyTransformation> transformations =
                                asymUnit.getPDBHeader().getBioAssemblies().get(bioAssemblyId).getTransforms();


                if (transformations == null || transformations.size() == 0) {
                        throw new StructureException("Could not load transformations to recreate biological assembly id " + bioAssemblyId + " of " + pdbId);
                }

                BiologicalAssemblyBuilder builder = new BiologicalAssemblyBuilder();

                // if we use mmcif or mmtf, then we need to pass useAsymIds=true
                boolean useAsymIds = false;
                if (filetype == StructureFiletype.CIF || filetype == StructureFiletype.BCIF || filetype == StructureFiletype.MMTF) {
                        useAsymIds = true;
                }
                return builder.rebuildQuaternaryStructure(asymUnit, transformations, useAsymIds, multiModel);
        }

        /**
         * Returns all biological assemblies for given PDB id.
         * @param pdbId
         * @param multiModel if true the output Structure will be a multi-model one with one transformId per model,
         * if false the outputStructure will be as the original with added chains with renamed asymIds (in the form originalAsymId_transformId and originalAuthId_transformId).
         * @return
         * @throws StructureException
         * @throws IOException
         * @since 5.0
         */
        public List<Structure> getBiologicalAssemblies(String pdbId, boolean multiModel) throws StructureException, IOException {
                List<Structure> assemblies = new ArrayList<>();

                boolean prevIsParseBioAssembly = getFileParsingParams().isParseBioAssembly();

                if (!getFileParsingParams().isParseBioAssembly()) {
                        getFileParsingParams().setParseBioAssembly(true);
                }

                Structure asymUnit = getStructureForPdbId(pdbId);
                getFileParsingParams().setParseBioAssembly(prevIsParseBioAssembly);

                if (asymUnit.getPDBHeader() == null || asymUnit.getPDBHeader().getBioAssemblies() == null) {
                        logger.info("No bioassembly information found for {}, returning asymmetric unit as the only biological assembly", pdbId);
                        assemblies.add(asymUnit);
                        return assemblies;
                }

                for (int bioAssemblyId : asymUnit.getPDBHeader().getBioAssemblies().keySet()) {
                        List<BiologicalAssemblyTransformation> transformations =
                                        asymUnit.getPDBHeader().getBioAssemblies().get(bioAssemblyId).getTransforms();

                        if (transformations == null || transformations.size() == 0) {
                                logger.info("Could not load transformations to recreate biological assembly id {} of {}. Assembly " +
                                                "id will be missing in biological assemblies.", bioAssemblyId, pdbId);
                                continue;
                        }

                        BiologicalAssemblyBuilder builder = new BiologicalAssemblyBuilder();

                        // if we use mmcif or mmtf, then we need to pass useAsymIds=true
                        boolean useAsymIds = false;
                        if (filetype == StructureFiletype.CIF || filetype == StructureFiletype.BCIF || filetype == StructureFiletype.MMTF) {
                                useAsymIds = true;
                        }
                        Structure s = builder.rebuildQuaternaryStructure(asymUnit, transformations, useAsymIds, multiModel);
                        assemblies.add(s);
                }
                return assemblies;
        }

        /**
         * Returns the path that contains the caching file for utility data, such as domain definitions.
         *
         * @return
         */
        public String getCachePath() {
                return cachePath;
        }

        public FileParsingParameters getFileParsingParams() {
                return params;
        }

        /**
         * Get the path that is used to cache PDB files.
         *
         * @return path to a directory
         */
        public String getPath() {
                return path;
        }

        /**
         * Request a Structure based on a <i>name</i>.
         *
         * <pre>
         *              Formal specification for how to specify the <i>name</i>:
         *
         *              name     := pdbID
         *                             | pdbID '.' chainID
         *                             | pdbID '.' range
         *                             | scopID
         *              range         := '('? range (',' range)? ')'?
         *                             | chainID
         *                             | chainID '_' resNum '-' resNum
         *              pdbID         := [1-9][a-zA-Z0-9]{3}
         *                             | PDB_[a-zA-Z0-9]{8}
         *              chainID       := [a-zA-Z0-9]
         *              scopID        := 'd' pdbID [a-z_][0-9_]
         *              resNum        := [-+]?[0-9]+[A-Za-z]?
         *
         *
         *              Example structures:
         *              1TIM                 #whole structure
         *              4HHB.C               #single chain
         *              4GCR.A_1-83          #one domain, by residue number
         *              3AA0.A,B             #two chains treated as one structure
         *              PDB_00001TIM         #whole structure (extended format)
         *              PDB_00004HHB.C       #single chain (extended format)
         *              PDB_00004GCR.A_1-83  #one domain, by residue number (extended format)
         *              PDB_00003AA0.A,B     #two chains treated as one structure (extended format)
         *              d2bq6a1              #scop domain
         * </pre>
         *
         * With the additional set of rules:
         *
         * <ul>
         * <li>If only a PDB code is provided, the whole structure will be return including ligands, but the first model
         * only (for NMR).
         * <li>Chain IDs are case sensitive, PDB ids are not. To specify a particular chain write as: 4hhb.A or 4HHB.A</li>
         * <li>To specify a SCOP domain write a scopId e.g. d2bq6a1. </li>
         * <li>URLs are accepted as well</li>
         * </ul>
         *
         * <p>Note that this method should not be used in StructureIdentifier
         * implementations to avoid circular calls.
         * @param name
         * @return a Structure object, or null if name appears improperly formated (eg too short, etc)
         * @throws IOException
         *             The PDB file cannot be cached due to IO errors
         * @throws StructureException
         *             The name appeared valid but did not correspond to a structure. Also thrown by some submethods upon
         *             errors, eg for poorly formatted subranges.
         */
        public Structure getStructure(String name) throws IOException, StructureException {
                StructureName structureName = new StructureName(name);
                return getStructure(structureName);
        }

        /**
         * Get the structure corresponding to the given {@link StructureIdentifier}.
         * Equivalent to calling {@link StructureIdentifier#loadStructure(AtomCache)}
         * followed by {@link StructureIdentifier#reduce(Structure)}.
         *
         * <p>Note that this method should not be used in StructureIdentifier
         * implementations to avoid circular calls.
         * @param strucId
         * @return
         * @throws IOException
         * @throws StructureException
         */
        public Structure getStructure(StructureIdentifier strucId) throws IOException, StructureException {
                Structure s = strucId.loadStructure(this);
                Structure r = strucId.reduce(s);
                r.setStructureIdentifier(strucId);
                return r;
        }

        /**
         * Returns the representation of a {@link ScopDomain} as a BioJava {@link Structure} object.
         *
         * @param domain
         *            a SCOP domain
         * @return a Structure object
         * @throws IOException
         * @throws StructureException
         */
        public Structure getStructureForDomain(ScopDomain domain) throws IOException, StructureException {
                return getStructureForDomain(domain, ScopFactory.getSCOP());
        }

        /**
         * Returns the representation of a {@link ScopDomain} as a BioJava {@link Structure} object.
         *
         * @param domain
         *            a SCOP domain
         * @param scopDatabase
         *            A {@link ScopDatabase} to use
         * @return a Structure object
         * @throws IOException
         * @throws StructureException
         */
        public Structure getStructureForDomain(ScopDomain domain, ScopDatabase scopDatabase) throws IOException,
                        StructureException {
                return getStructureForDomain(domain, scopDatabase, false);
        }

        /**
         * Returns the representation of a {@link ScopDomain} as a BioJava {@link Structure} object.
         *
         * @param domain
         *            a SCOP domain
         * @param scopDatabase
         *            A {@link ScopDatabase} to use
         * @param strictLigandHandling
         *            If set to false, hetero-atoms are included if and only if they belong to a chain to which the SCOP
         *            domain belongs; if set to true, hetero-atoms are included if and only if they are strictly within the
         *            definition (residue numbers) of the SCOP domain
         * @return a Structure object
         * @throws IOException
         * @throws StructureException
         */
        public Structure getStructureForDomain(ScopDomain domain, ScopDatabase scopDatabase, boolean strictLigandHandling)
                        throws IOException, StructureException {
                PdbId pdbId = domain.getPdbId();
                Structure fullStructure = getStructureForPdbId(pdbId);
                Structure structure = domain.reduce(fullStructure);

                // TODO It would be better to move all of this into the reduce method,
                // but that would require ligand handling properties in StructureIdentifiers

                // because ligands sometimes occur after TER records in PDB files, we may need to add some ligands back in
                // specifically, we add a ligand if and only if it occurs within the domain
                AtomPositionMap map = null;
                List<ResidueRangeAndLength> rrs = null;
                if (strictLigandHandling) {
                        map = new AtomPositionMap(StructureTools.getAllAtomArray(fullStructure), AtomPositionMap.ANYTHING_MATCHER);
                        rrs = ResidueRangeAndLength.parseMultiple(domain.getRanges(), map);
                }
                for (Chain chain : fullStructure.getNonPolyChains()) {
                        if (!structure.hasPdbChain(chain.getName())) {
                                continue; // we can't do anything with a chain our domain
                        }

                        Chain newChain;
                        if (!structure.hasNonPolyChain(chain.getId())) {
                                newChain = new ChainImpl();
                                newChain.setId(chain.getId());
                                newChain.setName(chain.getName());
                                newChain.setEntityInfo(chain.getEntityInfo());
                                structure.addChain(newChain);
                        } else {
                                newChain = structure.getNonPolyChain(chain.getId());
                        }

                        List<Group> ligands = StructureTools.filterLigands(chain.getAtomGroups());
                        for (Group group : ligands) {
                                boolean shouldContain = true;
                                if (strictLigandHandling) {
                                        shouldContain = false; // whether the ligand occurs within the domain
                                        for (ResidueRange rr : rrs) {
                                                if (rr.contains(group.getResidueNumber(), map)) {
                                                        shouldContain = true;
                                                }
                                        }
                                }
                                boolean alreadyContains = newChain.getAtomGroups().contains(group); // we don't want to add duplicate
                                                                                                                                                                        // ligands
                                if (shouldContain && !alreadyContains) {
                                        newChain.addGroup(group);
                                }
                        }
                }

                // build a more meaningful description for the new structure
                StringBuilder header = new StringBuilder();
                header.append(domain.getClassificationId());
                if (scopDatabase != null) {
                        int sf = domain.getSuperfamilyId();
                        ScopDescription description = scopDatabase.getScopDescriptionBySunid(sf);
                        if (description != null) {
                                header.append(" | ");
                                header.append(description.getDescription());
                        }
                }
                structure.getPDBHeader().setDescription(header.toString());

                return structure;
        }

        /**
         * Returns the representation of a {@link ScopDomain} as a BioJava {@link Structure} object.
         *
         * @param scopId
         *            a SCOP Id
         * @return a Structure object
         * @throws IOException
         * @throws StructureException
         */
        public Structure getStructureForDomain(String scopId) throws IOException, StructureException {
                return getStructureForDomain(scopId, ScopFactory.getSCOP());
        }

        /**
         * Returns the representation of a {@link ScopDomain} as a BioJava {@link Structure} object.
         *
         * @param scopId
         *            a SCOP Id
         * @param scopDatabase
         *            A {@link ScopDatabase} to use
         * @return a Structure object
         * @throws IOException
         * @throws StructureException
         */
        public Structure getStructureForDomain(String scopId, ScopDatabase scopDatabase) throws IOException,
                        StructureException {
                ScopDomain domain = scopDatabase.getDomainByScopID(scopId);
                return getStructureForDomain(domain, scopDatabase);
        }

        /**
         * set the location at which utility data should be cached.
         *
         * @param cachePath
         */
        public void setCachePath(String cachePath) {
                this.cachePath = cachePath;
        }

        public void setFileParsingParams(FileParsingParameters params) {
                this.params = params;
        }

        /**
         * <b>[Optional]</b> This method changes the behavior when obsolete entries
         * are requested. Current behaviors are:
         * <ul>
         * <li>{@link ObsoleteBehavior#THROW_EXCEPTION THROW_EXCEPTION}
         *   Throw a {@link StructureException} (the default)
         * <li>{@link ObsoleteBehavior#FETCH_OBSOLETE FETCH_OBSOLETE}
         *   Load the requested ID from the PDB's obsolete repository
         * <li>{@link ObsoleteBehavior#FETCH_CURRENT FETCH_CURRENT}
         *   Load the most recent version of the requested structure
         * </ul>
         *
         * <p>This setting may be silently ignored by implementations which do not have
         * access to the server to determine whether an entry is obsolete, such as
         * certain {@link FetchBehavior}s. Note that an obsolete entry may still be
         * returned even this is FETCH_CURRENT if the entry is found locally.
         *
         * @param behavior Whether to fetch obsolete records
         * @since 4.0.0
         */
        public void setObsoleteBehavior(ObsoleteBehavior behavior) {
                obsoleteBehavior = behavior;
        }

        /**
         * Returns how this instance deals with obsolete entries. Note that this
         * setting may be ignored by some implementations or in some situations,
         * such as certain {@link FetchBehavior}s.
         *
         * <p>For most implementations, the default value is
         * {@link ObsoleteBehavior#THROW_EXCEPTION THROW_EXCEPTION}.
         *
         * @return The ObsoleteBehavior
         * @since 4.0.0
         */
        public ObsoleteBehavior getObsoleteBehavior() {
                return obsoleteBehavior;
        }

        /**
         * Get the behavior for fetching files from the server
         * @return
         */
        public FetchBehavior getFetchBehavior() {
                return fetchBehavior;
        }

        /**
         * Set the behavior for fetching files from the server
         * @param fetchBehavior
         */
        public void setFetchBehavior(FetchBehavior fetchBehavior) {
                this.fetchBehavior = fetchBehavior;
        }

        /**
         * Set the path that is used to cache PDB files.
         *
         * @param path
         *            to a directory
         */
        public void setPath(String path) {
                this.path = FileDownloadUtils.expandUserHome(path);
        }

        /**
         * Returns the currently active file type that will be parsed.
         * @return a StructureFiletype
         */
        public StructureFiletype getFiletype() {
                return filetype;
        }

        /**
         * Set the file type that will be parsed.
         * @param filetype a StructureFiletype
         */
        public void setFiletype(StructureFiletype filetype) {
                this.filetype = filetype;
        }

        private boolean checkLoading(PdbId pdbId) {
                return currentlyLoading.contains(pdbId.getId());
        }

        /**
         * Returns a {@link Structure} corresponding to the CATH identifier supplied in {@code structureName}, using the the {@link CathDatabase}
         * at {@link CathFactory#getCathDatabase()}.
         */
        public Structure getStructureForCathDomain(StructureName structureName) throws IOException, StructureException {
                return getStructureForCathDomain(structureName, CathFactory.getCathDatabase());
        }

        /**
         * Returns a {@link Structure} corresponding to the CATH identifier supplied in {@code structureName}, using the specified {@link CathDatabase}.
         */
        public Structure getStructureForCathDomain(StructureName structureName, CathDatabase cathInstall) throws IOException, StructureException {
                CathDomain cathDomain = cathInstall.getDomainByCathId(structureName.getIdentifier());

                Structure s = getStructureForPdbId(cathDomain.getIdentifier());
                Structure n = cathDomain.reduce(s);

                // add the ligands of the chain...
                Chain newChain = n.getPolyChainByPDB(structureName.getChainId());
                List<Chain> origChains = s.getNonPolyChainsByPDB(structureName.getChainId());
                for (Chain origChain : origChains) {
                        List<Group> ligands = origChain.getAtomGroups();

                        for (Group g : ligands) {
                                if (!newChain.getAtomGroups().contains(g)) {
                                        newChain.addGroup(g);
                                }
                        }
                }

                return n;
        }

        protected void flagLoading(PdbId pdbId) {
                String id = pdbId.getId();
                if (!currentlyLoading.contains(id)) {
                        currentlyLoading.add(id);
                }
        }

        protected void flagLoadingFinished(PdbId pdbId) {
                currentlyLoading.remove(pdbId.getId());
        }

        /**
         * Loads a structure directly by PDB ID
         * @param id
         * @return
         * @throws IOException
         * @throws StructureException
         */
        public Structure getStructureForPdbId(String id) throws IOException, StructureException {
                if (id == null)
                        return null;
                return getStructureForPdbId(new PdbId(id));
        }
        /**
         * Loads a structure directly by PDB ID
         * @param pdbId
         * @return
         * @throws IOException
         */
        public Structure getStructureForPdbId(PdbId pdbId) throws IOException {
                if (pdbId == null)
                        return null;
                
                while (checkLoading(pdbId)) {
                        // waiting for loading to be finished...
                        try {
                                Thread.sleep(100);
                        } catch (InterruptedException e) {
                                logger.error(e.getMessage());
                        }
                }

                switch (filetype) {
                        case CIF:
                                logger.debug("loading from mmcif");
                                return loadStructureFromCifByPdbId(pdbId);
                        case BCIF:
                                logger.debug("loading from bcif");
                                return loadStructureFromBcifByPdbId(pdbId);
                        case PDB: default:
                                logger.debug("loading from pdb");
                                return loadStructureFromPdbByPdbId(pdbId);
                }
        }

        protected Structure loadStructureFromCifByPdbId(String pdbId) throws IOException {
                return loadStructureFromCifByPdbId(new PdbId(pdbId));
        }
        
        protected Structure loadStructureFromCifByPdbId(PdbId pdbId) throws IOException {
                logger.debug("Loading structure {} from mmCIF file {}.", pdbId, path);
                Structure s;
                flagLoading(pdbId);
                try {
                        CifFileReader reader = new CifFileReader(path);
                        reader.setFetchBehavior(fetchBehavior);
                        reader.setObsoleteBehavior(obsoleteBehavior);
                        reader.setFileParsingParameters(params);
                        s = reader.getStructureById(pdbId);
                } finally {
                        flagLoadingFinished(pdbId);
                }

                return s;
        }

        protected Structure loadStructureFromBcifByPdbId(String pdbId) throws IOException {
                return loadStructureFromBcifByPdbId(new PdbId(pdbId));
        }
        protected Structure loadStructureFromBcifByPdbId(PdbId pdbId) throws IOException {
                logger.debug("Loading structure {} from BinaryCIF file {}.", pdbId, path);
                Structure s;
                flagLoading(pdbId);
                try {
                        BcifFileReader reader = new BcifFileReader(path);
                        reader.setFetchBehavior(fetchBehavior);
                        reader.setObsoleteBehavior(obsoleteBehavior);
                        reader.setFileParsingParameters(params);
                        s = reader.getStructureById(pdbId);
                } finally {
                        flagLoadingFinished(pdbId);
                }

                return s;
        }

        protected Structure loadStructureFromPdbByPdbId(String pdbId) throws IOException {
                return loadStructureFromPdbByPdbId(new PdbId(pdbId));
        }

        protected Structure loadStructureFromPdbByPdbId(PdbId pdbId) throws IOException {
                logger.debug("Loading structure {} from PDB file {}.", pdbId, path);
                Structure s;
                flagLoading(pdbId);
                try {
                        PDBFileReader reader = new PDBFileReader(path);
                        reader.setFetchBehavior(fetchBehavior);
                        reader.setObsoleteBehavior(obsoleteBehavior);

                        reader.setFileParsingParameters(params);

                        s = reader.getStructureById(pdbId);
                } finally {
                        flagLoadingFinished(pdbId);
                }

                return s;
        }
}