X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fanalysis%2FSeqsetUtils.java;h=e2ed9b451ca94bd64da7c0a675835ad4c1ab205c;hb=841485c28403b4a8e00c9bddae9caa9411667b8a;hp=a0d0f0030b3540f26ec603b86340fb4563482a09;hpb=153dd62dc91da13ae732600e6ea55ddbe15eab39;p=jalview.git

diff --git a/src/jalview/analysis/SeqsetUtils.java b/src/jalview/analysis/SeqsetUtils.java
index a0d0f00..e2ed9b4 100755
--- a/src/jalview/analysis/SeqsetUtils.java
+++ b/src/jalview/analysis/SeqsetUtils.java
@@ -1,81 +1,91 @@
 /*
- * Jalview - A Sequence Alignment Editor and Viewer (Version 2.6)
- * Copyright (C) 2010 J Procter, AM Waterhouse, G Barton, M Clamp, S Searle
+ * Jalview - A Sequence Alignment Editor and Viewer ($$Version-Rel$$)
+ * Copyright (C) $$Year-Rel$$ The Jalview Authors
  * 
  * This file is part of Jalview.
  * 
  * Jalview 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, either version 3 of the License, or (at your option) any later version.
- * 
+ * as published by the Free Software Foundation, either version 3
+ * of the License, or (at your option) any later version.
+ *  
  * Jalview 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.
  * 
- * You should have received a copy of the GNU General Public License along with Jalview.  If not, see <http://www.gnu.org/licenses/>.
+ * You should have received a copy of the GNU General Public License
+ * along with Jalview.  If not, see <http://www.gnu.org/licenses/>.
+ * The Jalview Authors are detailed in the 'AUTHORS' file.
  */
 package jalview.analysis;
 
-import java.util.*;
+import jalview.bin.Cache;
+import jalview.datamodel.AlignmentAnnotation;
+import jalview.datamodel.HiddenMarkovModel;
+import jalview.datamodel.PDBEntry;
+import jalview.datamodel.Sequence;
+import jalview.datamodel.SequenceFeature;
+import jalview.datamodel.SequenceI;
 
-import jalview.datamodel.*;
+import java.util.ArrayList;
+import java.util.Enumeration;
+import java.util.HashMap;
+import java.util.Hashtable;
+import java.util.Iterator;
+import java.util.List;
+import java.util.Map;
+import java.util.Objects;
+import java.util.Optional;
+import java.util.Vector;
+import static java.lang.String.format;
 
-/**
- * <p>
- * Title:
- * </p>
- * 
- * <p>
- * Description:
- * </p>
- * 
- * <p>
- * Copyright: Copyright (c) 2004
- * </p>
- * 
- * <p>
- * Company: Dundee University
- * </p>
- * 
- * @author not attributable
- * @version 1.0
- */
 public class SeqsetUtils
 {
+  public static class SequenceInfo {
+    private String name;
+    private int start;
+    private int end;
+    private Optional<String> description = Optional.empty();
+    private Optional<List<SequenceFeature>> features = Optional.empty();
+    private Optional<List<PDBEntry>> pdbId = Optional.empty();
+    private Optional<SequenceI> dataset = Optional.empty();
+    private Optional<HiddenMarkovModel> hmm = Optional.empty();
+    private Optional<AlignmentAnnotation[]> searchScores = Optional.empty();
+
+    private SequenceInfo(String name, int start, int end) {
+      this.name = name;
+      this.start = start;
+      this.end = end;
+    }
+  }
 
   /**
    * Store essential properties of a sequence in a hashtable for later recovery
-   * Keys are Name, Start, End, SeqFeatures, PdbId
+   * Keys are Name, Start, End, SeqFeatures, PdbId, HMM
    * 
    * @param seq
    *          SequenceI
    * @return Hashtable
    */
-  public static Hashtable SeqCharacterHash(SequenceI seq)
+  public static SequenceInfo SeqCharacterHash(SequenceI seq)
   {
-    Hashtable sqinfo = new Hashtable();
-    sqinfo.put("Name", seq.getName());
-    sqinfo.put("Start", new Integer(seq.getStart()));
-    sqinfo.put("End", new Integer(seq.getEnd()));
-    if (seq.getDescription() != null)
+    SequenceInfo sqinfo = new SequenceInfo(seq.getName(), seq.getStart(), seq.getEnd());
+    sqinfo.description = Optional.ofNullable(seq.getDescription());
+    sqinfo.dataset = Optional.ofNullable(seq.getDatasetSequence());
+    if (!sqinfo.dataset.isPresent())
     {
-      sqinfo.put("Description", seq.getDescription());
+      ArrayList<SequenceFeature> feats = new ArrayList<>(
+          seq.getFeatures().getAllFeatures());
+      sqinfo.features = Optional.of(feats);
+      sqinfo.pdbId = Optional.of(Objects.requireNonNullElse(
+          seq.getAllPDBEntries(), new ArrayList<>()));
     }
-    Vector sfeat = new Vector();
-    jalview.datamodel.SequenceFeature[] sfarray = seq.getSequenceFeatures();
-    if (sfarray != null && sfarray.length > 0)
+    if (seq.hasHMMProfile())
     {
-      for (int i = 0; i < sfarray.length; i++)
-      {
-        sfeat.addElement(sfarray[i]);
-      }
+      sqinfo.hmm = Optional.of(seq.getHMM());
     }
-    sqinfo.put("SeqFeatures", sfeat);
-    sqinfo.put("PdbId", (seq.getPDBId() != null) ? seq.getPDBId()
-            : new Vector());
-    sqinfo.put("datasetSequence", (seq.getDatasetSequence() != null) ? seq
-            .getDatasetSequence() : new Sequence("THISISAPLACEHOLDER", ""));
+    sqinfo.searchScores = Optional.ofNullable(seq.getAnnotation("Search Scores"));
     return sqinfo;
   }
 
@@ -89,60 +99,44 @@ public class SeqsetUtils
    *          Hashtable
    * @return boolean true if name was not updated from sqinfo Name entry
    */
-  public static boolean SeqCharacterUnhash(SequenceI sq, Hashtable sqinfo)
+  public static boolean SeqCharacterUnhash(SequenceI sq, SequenceInfo sqinfo)
   {
-    boolean namePresent = true;
     if (sqinfo == null)
     {
       return false;
     }
-    String oldname = (String) sqinfo.get("Name");
-    Integer start = (Integer) sqinfo.get("Start");
-    Integer end = (Integer) sqinfo.get("End");
-    Vector sfeatures = (Vector) sqinfo.get("SeqFeatures");
-    Vector pdbid = (Vector) sqinfo.get("PdbId");
-    String description = (String) sqinfo.get("Description");
-    Sequence seqds = (Sequence) sqinfo.get("datasetSequence");
-    if (oldname == null)
+    if (sqinfo.name != null)
     {
-      namePresent = false;
+      sq.setName(sqinfo.name);
     }
-    else
+    sq.setStart(sqinfo.start);
+    sq.setEnd(sqinfo.end);
+    if (sqinfo.pdbId.isPresent() && !sqinfo.pdbId.get().isEmpty())
+      sq.setPDBId(new Vector<>(sqinfo.pdbId.get()));
+    if (sqinfo.features.isPresent() && !sqinfo.features.get().isEmpty())
+      sq.setSequenceFeatures(sqinfo.features.get());
+    if (sqinfo.description.isPresent())
+      sq.setDescription(sqinfo.description.get());
+    if (sqinfo.dataset.isPresent())
     {
-      sq.setName(oldname);
-    }
-    if (pdbid != null && pdbid.size() > 0)
-    {
-      sq.setPDBId(pdbid);
-    }
-
-    if ((start != null) && (end != null))
-    {
-      sq.setStart(start.intValue());
-      sq.setEnd(end.intValue());
-    }
-
-    if ((sfeatures != null) && (sfeatures.size() > 0))
-    {
-      SequenceFeature[] sfarray = new SequenceFeature[sfeatures.size()];
-      for (int is = 0, isize = sfeatures.size(); is < isize; is++)
+      if (sqinfo.features.isPresent())
       {
-        sfarray[is] = (SequenceFeature) sfeatures.elementAt(is);
+        Cache.log.warn("Setting dataset sequence for a sequence which has " +
+            "sequence features. Dataset sequence features will not be visible.");
+        assert false;
       }
-      sq.setSequenceFeatures(sfarray);
-    }
-    if (description != null)
-    {
-      sq.setDescription(description);
+      sq.setDatasetSequence(sqinfo.dataset.get());
     }
-    if ((seqds != null)
-            && !(seqds.getName().equals("THISISAPLACEHOLDER") && seqds
-                    .getLength() == 0))
+    if (sqinfo.hmm.isPresent())
+      sq.setHMM(new HiddenMarkovModel(sqinfo.hmm.get(), sq));
+    if (sqinfo.searchScores.isPresent())
     {
-      sq.setDatasetSequence(seqds);
+      for (AlignmentAnnotation score : sqinfo.searchScores.get())
+      {
+        sq.addAlignmentAnnotation(score);
+      }
     }
-
-    return namePresent;
+    return sqinfo.name != null;
   }
 
   /**
@@ -155,7 +149,7 @@ public class SeqsetUtils
    */
   public static String unique_name(int i)
   {
-    return new String("Sequence" + i);
+    return String.format("Sequence%d", i);
   }
 
   /**
@@ -172,12 +166,12 @@ public class SeqsetUtils
    * @see deuniquify to recover original names (and properties) for renamed
    *      sequences
    */
-  public static Hashtable uniquify(SequenceI[] sequences,
+  public static Map<String, SequenceInfo> uniquify(SequenceI[] sequences,
           boolean write_names)
   {
     // Generate a safely named sequence set and a hash to recover the sequence
     // names
-    Hashtable map = new Hashtable();
+    HashMap<String, SequenceInfo> map = new HashMap<>();
     // String[] un_names = new String[sequences.length];
 
     for (int i = 0; i < sequences.length; i++)
@@ -205,7 +199,8 @@ public class SeqsetUtils
    *          SequenceI[]
    * @return boolean
    */
-  public static boolean deuniquify(Hashtable map, SequenceI[] sequences)
+  public static boolean deuniquify(Map<String, SequenceInfo> map,
+      SequenceI[] sequences)
   {
     return deuniquify(map, sequences, true);
   }
@@ -224,26 +219,25 @@ public class SeqsetUtils
    *          map.
    * @return boolean
    */
-  public static boolean deuniquify(Hashtable map, SequenceI[] sequences,
-          boolean quiet)
+  public static boolean deuniquify(Map<String, SequenceInfo> map,
+      SequenceI[] sequences, boolean quiet)
   {
     jalview.analysis.SequenceIdMatcher matcher = new SequenceIdMatcher(
             sequences);
     SequenceI msq = null;
-    Enumeration keys = map.keys();
-    Vector unmatched = new Vector();
+    Iterator<String> keys = map.keySet().iterator();
+    Vector<SequenceI> unmatched = new Vector<>();
     for (int i = 0, j = sequences.length; i < j; i++)
     {
       unmatched.addElement(sequences[i]);
     }
-    while (keys.hasMoreElements())
+    while (keys.hasNext())
     {
-      Object key = keys.nextElement();
-      if (key instanceof String)
-      {
+      String key = keys.next();
+      try {
         if ((msq = matcher.findIdMatch((String) key)) != null)
         {
-          Hashtable sqinfo = (Hashtable) map.get(key);
+          SequenceInfo sqinfo = map.get(key);
           unmatched.removeElement(msq);
           SeqCharacterUnhash(msq, sqinfo);
         }
@@ -251,20 +245,27 @@ public class SeqsetUtils
         {
           if (!quiet)
           {
-            System.err.println("Can't find '" + ((String) key)
-                    + "' in uniquified alignment");
+            Cache.log.warn(format("Can't find '%s' in uniquified alignment",
+                key));
           }
         }
+      } catch (ClassCastException ccastex) {
+        if (!quiet)
+        {
+          Cache.log.error("Unexpected object in SeqSet map : "+ key.getClass());
+        }
       }
     }
     if (unmatched.size() > 0 && !quiet)
     {
-      System.err.println("Did not find matches for :");
-      for (Enumeration i = unmatched.elements(); i.hasMoreElements(); System.out
-              .println(((SequenceI) i.nextElement()).getName()))
+      StringBuilder sb = new StringBuilder("Did not find match for sequences: ");
+      Enumeration<SequenceI> i = unmatched.elements();
+      sb.append(i.nextElement().getName());
+      for (; i.hasMoreElements();)
       {
-        ;
+        sb.append(", " + i.nextElement().getName());
       }
+      Cache.log.warn(sb.toString());
       return false;
     }
 
@@ -287,8 +288,8 @@ public class SeqsetUtils
     for (int i = 0, j = sequences.length; i < j; i++)
     {
       String tempseq = jalview.analysis.AlignSeq.extractGaps(
-              jalview.util.Comparison.GapChars, sequences[i]
-                      .getSequenceAsString());
+              jalview.util.Comparison.GapChars,
+              sequences[i].getSequenceAsString());
 
       if (tempseq.length() == 0)
       {