X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fanalysis%2FSequenceIdMatcher.java;h=a51f2f21e5f74d3fb4e67ebc7b66e5bc73a50a14;hb=838e4f91d4a53dd315640dbc9ff6ef7a815ee576;hp=0046cf565ce82b82a4303ba12d03ab0d022e4fcc;hpb=153dd62dc91da13ae732600e6ea55ddbe15eab39;p=jalview.git

diff --git a/src/jalview/analysis/SequenceIdMatcher.java b/src/jalview/analysis/SequenceIdMatcher.java
index 0046cf5..a51f2f2 100755
--- a/src/jalview/analysis/SequenceIdMatcher.java
+++ b/src/jalview/analysis/SequenceIdMatcher.java
@@ -1,70 +1,78 @@
 /*
- * 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 2.9.0b1)
+ * Copyright (C) 2015 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.datamodel.DBRefEntry;
+import jalview.datamodel.SequenceI;
 
-import jalview.datamodel.*;
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Vector;
 
 /**
- * <p>
- * Title:
- * </p>
- * SequenceIdMatcher
- * <p>
- * Description:
- * </p>
- * Routine which does approximate Sequence Id resolution by name using string
+ * Routines for approximate Sequence Id resolution by name using string
  * containment (on word boundaries) rather than equivalence. It also attempts to
  * resolve ties where no exact match is available by picking the the id closest
  * to the query.
- * <p>
- * Copyright: Copyright (c) 2004
- * </p>
- * 
- * <p>
- * Company: Dundee University
- * </p>
- * 
- * @author not attributable
- * @version 1.0
  */
 public class SequenceIdMatcher
 {
-  private Hashtable names;
+  private HashMap<SeqIdName, SequenceI> names;
+
+  public SequenceIdMatcher(List<SequenceI> seqs)
+  {
+    names = new HashMap<SeqIdName, SequenceI>();
+    addAll(seqs);
+  }
 
-  public SequenceIdMatcher(SequenceI[] seqs)
+  /**
+   * add more sequences to this matcher - also used by the constructor
+   * 
+   * @param seqs
+   */
+  public void addAll(List<SequenceI> seqs)
   {
-    names = new Hashtable();
-    for (int i = 0; i < seqs.length; i++)
+    for (SequenceI seq : seqs)
     {
-      names.put(new SeqIdName(seqs[i].getName()), seqs[i]);
+      // TODO: deal with ID collisions - SequenceI should be appended to list
+      // associated with this key.
+      names.put(new SeqIdName(seq.getDisplayId(true)), seq);
+      SequenceI dbseq = seq;
+      while (dbseq.getDatasetSequence() != null)
+      {
+        dbseq = dbseq.getDatasetSequence();
+      }
       // add in any interesting identifiers
-      if (seqs[i].getDBRef() != null)
+      if (dbseq.getDBRef() != null)
       {
-        DBRefEntry dbr[] = seqs[i].getDBRef();
+        DBRefEntry dbr[] = dbseq.getDBRef();
         SeqIdName sid = null;
         for (int r = 0; r < dbr.length; r++)
         {
           sid = new SeqIdName(dbr[r].getAccessionId());
-          if (!names.contains(sid))
+          if (!names.containsKey(sid))
           {
-            names.put(sid, seqs[i]);
+            names.put(sid, seq);
           }
         }
       }
@@ -72,42 +80,84 @@ public class SequenceIdMatcher
   }
 
   /**
+   * convenience method to make a matcher from concrete array
+   * 
+   * @param sequences
+   */
+  public SequenceIdMatcher(SequenceI[] sequences)
+  {
+    this(Arrays.asList(sequences));
+  }
+
+  /**
    * returns the closest SequenceI in matches to SeqIdName and returns all the
    * matches to the names hash.
    * 
    * @param candName
    *          SeqIdName
    * @param matches
-   *          Vector of SequenceI objects
+   *          List of SequenceI objects
    * @return SequenceI closest SequenceI to SeqIdName
    */
-  private SequenceI pickbestMatch(SeqIdName candName, Vector matches)
+  private SequenceI pickbestMatch(SeqIdName candName,
+          List<SequenceI> matches)
+  {
+    List<SequenceI> st = pickbestMatches(candName, matches);
+    return st == null || st.size() == 0 ? null : st.get(0);
+  }
+
+  /**
+   * returns the closest SequenceI in matches to SeqIdName and returns all the
+   * matches to the names hash.
+   * 
+   * @param candName
+   *          SeqIdName
+   * @param matches
+   *          Vector of SequenceI objects
+   * @return Object[] { SequenceI closest SequenceI to SeqIdName, SequenceI[]
+   *         ties }
+   */
+  private List<SequenceI> pickbestMatches(SeqIdName candName,
+          List<SequenceI> matches)
   {
-    SequenceI match = null;
+    ArrayList<SequenceI> best = new ArrayList<SequenceI>();
     if (candName == null || matches == null || matches.size() == 0)
     {
       return null;
     }
-    match = (SequenceI) matches.elementAt(0);
-    matches.removeElementAt(0);
+    SequenceI match = matches.remove(0);
+    best.add(match);
     names.put(new SeqIdName(match.getName()), match);
     int matchlen = match.getName().length();
     int namlen = candName.id.length();
     while (matches.size() > 0)
     {
       // look through for a better one.
-      SequenceI cand = (SequenceI) matches.elementAt(0);
+      SequenceI cand = matches.remove(0);
       names.put(new SeqIdName(cand.getName()), cand);
-      int candlen = cand.getName().length();
+      int q, w, candlen = cand.getName().length();
       // keep the one with an id 'closer' to the given seqnam string
-      if (Math.abs(matchlen - namlen) > Math.abs(candlen - namlen)
+      if ((q = Math.abs(matchlen - namlen)) > (w = Math.abs(candlen
+              - namlen))
               && candlen > matchlen)
       {
+        best.clear();
         match = cand;
         matchlen = candlen;
+        best.add(match);
       }
+      if (q == w && candlen == matchlen)
+      {
+        // record any ties
+        best.add(cand);
+      }
+    }
+    if (best.size() == 0)
+    {
+      return null;
     }
-    return match;
+    ;
+    return best;
   }
 
   /**
@@ -117,19 +167,38 @@ public class SequenceIdMatcher
    *          SequenceI
    * @return SequenceI
    */
-  SequenceI findIdMatch(SequenceI seq)
+  public SequenceI findIdMatch(SequenceI seq)
   {
     SeqIdName nam = new SeqIdName(seq.getName());
     return findIdMatch(nam);
   }
 
-  SequenceI findIdMatch(String seqnam)
+  public SequenceI findIdMatch(String seqnam)
   {
     SeqIdName nam = new SeqIdName(seqnam);
     return findIdMatch(nam);
   }
 
   /**
+   * Find all matches for a given sequence name.
+   * 
+   * @param seqnam
+   *          string to query Matcher with.
+   * @return a new array or (possibly) null
+   */
+  public SequenceI[] findAllIdMatches(String seqnam)
+  {
+
+    SeqIdName nam = new SeqIdName(seqnam);
+    List<SequenceI> m = findAllIdMatches(nam);
+    if (m != null)
+    {
+      return m.toArray(new SequenceI[m.size()]);
+    }
+    return null;
+  }
+
+  /**
    * findIdMatch
    * 
    * Return pointers to sequences (or sequence object containers) which have
@@ -139,7 +208,7 @@ public class SequenceIdMatcher
    *          SequenceI[]
    * @return SequenceI[]
    */
-  SequenceI[] findIdMatch(SequenceI[] seqs)
+  public SequenceI[] findIdMatch(SequenceI[] seqs)
   {
     SequenceI[] namedseqs = null;
     int i = 0;
@@ -184,6 +253,25 @@ public class SequenceIdMatcher
     return pickbestMatch(nam, matches);
   }
 
+  /**
+   * core findIdMatch search method for finding all equivalent matches
+   * 
+   * @param nam
+   *          SeqIdName
+   * @return SequenceI[]
+   */
+  private List<SequenceI> findAllIdMatches(
+          jalview.analysis.SequenceIdMatcher.SeqIdName nam)
+  {
+    ArrayList<SequenceI> matches = new ArrayList<SequenceI>();
+    while (names.containsKey(nam))
+    {
+      matches.add(names.remove(nam));
+    }
+    List<SequenceI> r = pickbestMatches(nam, matches);
+    return r;
+  }
+
   private class SeqIdName
   {
     String id;
@@ -200,14 +288,20 @@ public class SequenceIdMatcher
       }
     }
 
+    @Override
     public int hashCode()
     {
       return ((id.length() >= 4) ? id.substring(0, 4).hashCode() : id
               .hashCode());
     }
 
+    @Override
     public boolean equals(Object s)
     {
+      if (s == null)
+      {
+        return false;
+      }
       if (s instanceof SeqIdName)
       {
         return this.equals((SeqIdName) s);
@@ -229,7 +323,8 @@ public class SequenceIdMatcher
      * arbritrarily extended sequence id's (like portions of an aligned set of
      * repeats from one sequence)
      */
-    private String WORD_SEP = "~. |#\\/<>!\"£$%^*)}[@',?_";
+    private String WORD_SEP = "~. |#\\/<>!\"" + ((char) 0x00A4)
+            + "$%^*)}[@',?_";
 
     /**
      * matches if one ID properly contains another at a whitespace boundary.
@@ -242,6 +337,8 @@ public class SequenceIdMatcher
      */
     public boolean equals(SeqIdName s)
     {
+      // TODO: JAL-732 patch for cases when name includes a list of IDs, and the
+      // match contains one ID flanked
       if (id.length() > s.id.length())
       {
         return id.startsWith(s.id) ? (WORD_SEP.indexOf(id.charAt(s.id