package jalview.datamodel;
import jalview.analysis.AlignSeq;
-import jalview.api.DBRefEntryI;
import jalview.datamodel.features.SequenceFeatures;
import jalview.datamodel.features.SequenceFeaturesI;
import jalview.util.Comparison;
/**
*
- * Implements the SequenceI interface for a char[] based sequence object.
- *
- * @author $author$
- * @version $Revision$
+ * Implements the SequenceI interface for a char[] based sequence object
*/
public class Sequence extends ASequence implements SequenceI
{
+
+ /**
+ * A subclass that gives us access to modCount, which tracks whether there
+ * have been any changes. We use this to update
+ *
+ * @author hansonr
+ *
+ * @param <T>
+ */
+ @SuppressWarnings("serial")
+ public class DBModList<T> extends ArrayList<DBRefEntry>
+ {
+
+ protected int getModCount()
+ {
+ return modCount;
+ }
+
+ }
+
SequenceI datasetSequence;
- String name;
+ private String name;
private char[] sequence;
- String description;
+ private String description;
- int start;
+ private int start;
- int end;
+ private int end;
- Vector<PDBEntry> pdbIds;
+ private Vector<PDBEntry> pdbIds;
- String vamsasId;
+ private String vamsasId;
- DBRefEntry[] dbrefs;
+ private DBModList<DBRefEntry> dbrefs; // controlled access
- RNA rna;
+ /**
+ * a flag to let us know that elements have changed in dbrefs
+ *
+ * @author Bob Hanson
+ */
+ private int refModCount = 0;
+
+ private RNA rna;
/**
* This annotation is displayed below the alignment but the positions are tied
*
* TODO: change to List<>
*/
- Vector<AlignmentAnnotation> annotation;
+ private Vector<AlignmentAnnotation> annotation;
private SequenceFeaturesI sequenceFeatureStore;
{
setDatasetSequence(seq.getDatasetSequence());
}
-
+
/*
* only copy DBRefs and seqfeatures if we really are a dataset sequence
*/
if (datasetSequence == null)
{
- if (seq.getDBRefs() != null)
+ List<DBRefEntry> dbr = seq.getDBRefs();
+ if (dbr != null)
{
- DBRefEntry[] dbr = seq.getDBRefs();
- for (int i = 0; i < dbr.length; i++)
+ for (int i = 0, n = dbr.size(); i < n; i++)
{
- addDBRef(new DBRefEntry(dbr[i]));
+ addDBRef(new DBRefEntry(dbr.get(i)));
}
}
{
if (sf.getType() == null)
{
- System.err.println("SequenceFeature type may not be null: "
- + sf.toString());
+ System.err.println(
+ "SequenceFeature type may not be null: " + sf.toString());
return false;
}
@Override
public Vector<PDBEntry> getAllPDBEntries()
{
- return pdbIds == null ? new Vector<PDBEntry>() : pdbIds;
+ return pdbIds == null ? new Vector<>() : pdbIds;
}
/**
public char[] getSequence()
{
// return sequence;
- return sequence == null ? null : Arrays.copyOf(sequence,
- sequence.length);
+ return sequence == null ? null
+ : Arrays.copyOf(sequence, sequence.length);
}
/*
public void setGeneLoci(String speciesId, String assemblyId,
String chromosomeId, MapList map)
{
- addDBRef(new DBRefEntry(speciesId, assemblyId, DBRefEntry.CHROMOSOME
- + ":" + chromosomeId, new Mapping(map)));
+ addDBRef(new GeneLocus(speciesId, assemblyId, chromosomeId,
+ new Mapping(map)));
}
/**
@Override
public GeneLociI getGeneLoci()
{
- DBRefEntry[] refs = getDBRefs();
+ List<DBRefEntry> refs = getDBRefs();
if (refs != null)
{
for (final DBRefEntry ref : refs)
{
- if (ref.isChromosome())
+ if (ref instanceof GeneLociI)
{
- return new GeneLociI()
- {
- @Override
- public String getSpeciesId()
- {
- return ref.getSource();
- }
-
- @Override
- public String getAssemblyId()
- {
- return ref.getVersion();
- }
-
- @Override
- public String getChromosomeId()
- {
- // strip off "chromosome:" prefix to chrId
- return ref.getAccessionId().substring(
- DBRefEntry.CHROMOSOME.length() + 1);
- }
-
- @Override
- public MapList getMap()
- {
- return ref.getMap().getMap();
- }
- };
+ return (GeneLociI) ref;
}
}
}
* preserve end residue column provided cursor was valid
*/
int endColumn = isValidCursor(cursor) ? cursor.lastColumnPosition : 0;
+
if (residuePos == this.end)
{
endColumn = column;
* @param curs
* @return
*/
- protected int findIndex(int pos, SequenceCursor curs)
+ protected int findIndex(final int pos, SequenceCursor curs)
{
if (!isValidCursor(curs))
{
/*
* move left or right to find pos from hint.position
*/
- int col = curs.columnPosition - 1; // convert from base 1 to 0-based array
- // index
+ int col = curs.columnPosition - 1; // convert from base 1 to base 0
int newPos = curs.residuePosition;
int delta = newPos > pos ? -1 : 1;
while (newPos != pos)
{
col += delta; // shift one column left or right
- if (col < 0 || col == sequence.length)
+ if (col < 0)
+ {
+ break;
+ }
+ if (col == sequence.length)
{
+ col--; // return last column if we failed to reach pos
break;
}
if (!Comparison.isGap(sequence[col]))
}
col++; // convert back to base 1
- updateCursor(pos, col, curs.firstColumnPosition);
+
+ /*
+ * only update cursor if we found the target position
+ */
+ if (newPos == pos)
+ {
+ updateCursor(pos, col, curs.firstColumnPosition);
+ }
return col;
}
{
return findPosition(column + 1, cursor);
}
-
+
// TODO recode this more naturally i.e. count residues only
// as they are found, not 'in anticipation'
* {@inheritDoc}
*/
@Override
- public Range findPositions(int fromColumn, int toColumn)
+ public ContiguousI findPositions(int fromColumn, int toColumn)
{
if (toColumn < fromColumn || fromColumn < 1)
{
{
ArrayList<int[]> map = new ArrayList<>();
int lastj = -1, j = 0;
- int pos = start;
+ // int pos = start;
int seqlen = sequence.length;
while ((j < seqlen))
{
{
BitSet map = new BitSet();
int lastj = -1, j = 0;
- int pos = start;
+ // int pos = start;
int seqlen = sequence.length;
while ((j < seqlen))
{
boolean createNewDs = false;
// TODO: take a (second look) at the dataset creation validation method for
// the very large sequence case
+
int startIndex = findIndex(start) - 1;
int endIndex = findIndex(end) - 1;
int startDeleteColumn = -1; // for dataset sequence deletions
int deleteCount = 0;
- for (int s = i; s < j; s++)
+ for (int s = i; s < j && s < sequence.length; s++)
{
if (Comparison.isGap(sequence[s]))
{
vamsasId = id;
}
+ @Deprecated
@Override
- public void setDBRefs(DBRefEntry[] dbref)
+ public void setDBRefs(DBModList<DBRefEntry> newDBrefs)
{
if (dbrefs == null && datasetSequence != null
&& this != datasetSequence)
{
- datasetSequence.setDBRefs(dbref);
+ datasetSequence.setDBRefs(newDBrefs);
return;
}
- dbrefs = dbref;
- if (dbrefs != null)
- {
- DBRefUtils.ensurePrimaries(this);
- }
+ dbrefs = newDBrefs;
+ refModCount = 0;
}
@Override
- public DBRefEntry[] getDBRefs()
+ public DBModList<DBRefEntry> getDBRefs()
{
if (dbrefs == null && datasetSequence != null
&& this != datasetSequence)
if (dbrefs == null)
{
- dbrefs = new DBRefEntry[0];
+ dbrefs = new DBModList<>();
}
- for (DBRefEntryI dbr : dbrefs)
+ for (int ib = 0, nb = dbrefs.size(); ib < nb; ib++)
{
- if (dbr.updateFrom(entry))
+ if (dbrefs.get(ib).updateFrom(entry))
{
/*
* found a dbref that either matched, or could be
}
}
- /*
- * extend the array to make room for one more
- */
- // TODO use an ArrayList instead
- int j = dbrefs.length;
- DBRefEntry[] temp = new DBRefEntry[j + 1];
- System.arraycopy(dbrefs, 0, temp, 0, j);
- temp[temp.length - 1] = entry;
-
- dbrefs = temp;
+ // /// BH OUCH!
+ // /*
+ // * extend the array to make room for one more
+ // */
+ // // TODO use an ArrayList instead
+ // int j = dbrefs.length;
+ // List<DBRefEntry> temp = new DBRefEntry[j + 1];
+ // System.arraycopy(dbrefs, 0, temp, 0, j);
+ // temp[temp.length - 1] = entry;
+ //
+ // dbrefs = temp;
- DBRefUtils.ensurePrimaries(this);
+ dbrefs.add(entry);
}
@Override
private int _seqhash = 0;
+ private List<DBRefEntry> primaryRefs;
+
/**
* Answers false if the sequence is more than 85% nucleotide (ACGTU), else
* true
_isNa = Comparison.isNucleotide(this);
}
return !_isNa;
- };
+ }
/*
* (non-Javadoc)
// TODO: could merge DBRefs
return datasetSequence.updatePDBIds();
}
- if (dbrefs == null || dbrefs.length == 0)
+ if (dbrefs == null || dbrefs.size() == 0)
{
return false;
}
boolean added = false;
- for (DBRefEntry dbr : dbrefs)
+ for (int ib = 0, nb = dbrefs.size(); ib < nb; ib++)
{
+ DBRefEntry dbr = dbrefs.get(ib);
if (DBRefSource.PDB.equals(dbr.getSource()))
{
/*
List<SequenceFeature> sfs = entry.getSequenceFeatures();
for (SequenceFeature feature : sfs)
{
- SequenceFeature sf[] = (mp != null) ? mp.locateFeature(feature)
- : new SequenceFeature[] { new SequenceFeature(feature) };
+ SequenceFeature sf[] = (mp != null) ? mp.locateFeature(feature)
+ : new SequenceFeature[]
+ { new SequenceFeature(feature) };
if (sf != null)
{
for (int sfi = 0; sfi < sf.length; sfi++)
}
}
// transfer database references
- DBRefEntry[] entryRefs = entry.getDBRefs();
+ List<DBRefEntry> entryRefs = entry.getDBRefs();
if (entryRefs != null)
{
- for (int r = 0; r < entryRefs.length; r++)
+ for (int r = 0, n = entryRefs.size(); r < n; r++)
{
- DBRefEntry newref = new DBRefEntry(entryRefs[r]);
+ DBRefEntry newref = new DBRefEntry(entryRefs.get(r));
if (newref.getMap() != null && mp != null)
{
// remap ref using our local mapping
return null;
}
+ private List<DBRefEntry> tmpList;
+
@Override
public List<DBRefEntry> getPrimaryDBRefs()
{
{
return datasetSequence.getPrimaryDBRefs();
}
- if (dbrefs == null || dbrefs.length == 0)
+ if (dbrefs == null || dbrefs.size() == 0)
{
return Collections.emptyList();
}
synchronized (dbrefs)
{
- List<DBRefEntry> primaries = new ArrayList<>();
- DBRefEntry[] tmp = new DBRefEntry[1];
- for (DBRefEntry ref : dbrefs)
+ if (refModCount == dbrefs.getModCount() && primaryRefs != null)
{
+ return primaryRefs; // no changes
+ }
+ refModCount = dbrefs.getModCount();
+ List<DBRefEntry> primaries = (primaryRefs == null
+ ? (primaryRefs = new ArrayList<>())
+ : primaryRefs);
+ primaries.clear();
+ if (tmpList == null)
+ {
+ tmpList = new ArrayList<>();
+ tmpList.add(null); // for replacement
+ }
+ for (int i = 0, n = dbrefs.size(); i < n; i++)
+ {
+ DBRefEntry ref = dbrefs.get(i);
if (!ref.isPrimaryCandidate())
{
continue;
}
}
// whilst it looks like it is a primary ref, we also sanity check type
- if (DBRefUtils.getCanonicalName(DBRefSource.PDB)
- .equals(DBRefUtils.getCanonicalName(ref.getSource())))
+ if (DBRefSource.PDB_CANONICAL_NAME
+ .equals(ref.getCanonicalSourceName()))
{
// PDB dbrefs imply there should be a PDBEntry associated
// TODO: tighten PDB dbrefs
// handle on the PDBEntry, and a real mapping between sequence and
// extracted sequence from PDB file
PDBEntry pdbentry = getPDBEntry(ref.getAccessionId());
- if (pdbentry != null && pdbentry.getFile() != null)
+ if (pdbentry == null || pdbentry.getFile() == null)
{
- primaries.add(ref);
+ continue;
}
- continue;
}
- // check standard protein or dna sources
- tmp[0] = ref;
- DBRefEntry[] res = DBRefUtils.selectDbRefs(!isProtein(), tmp);
- if (res != null && res[0] == tmp[0])
+ else
{
- primaries.add(ref);
- continue;
+ // check standard protein or dna sources
+ tmpList.set(0, ref);
+ List<DBRefEntry> res = DBRefUtils.selectDbRefs(!isProtein(),
+ tmpList);
+ if (res == null || res.get(0) != tmpList.get(0))
+ {
+ continue;
+ }
}
+ primaries.add(ref);
}
+
+ // version must be not null, as otherwise it will not be a candidate,
+ // above
+ DBRefUtils.ensurePrimaries(this, primaries);
return primaries;
}
}
List<SequenceFeature> result = getFeatures().findFeatures(startPos,
endPos, types);
+ if (datasetSequence != null)
+ {
+ result = datasetSequence.getFeatures().findFeatures(startPos, endPos,
+ types);
+ }
+ else
+ {
+ result = sequenceFeatureStore.findFeatures(startPos, endPos, types);
+ }
/*
* if end column is gapped, endPos may be to the right,
git://source.jalview.org / jalview.git / blobdiff