X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Futil%2FComparison.java;h=aa6bcd45e7d659507fe5a58590347ec1a6d5ab14;hb=refs%2Fheads%2Fpatch%2FJAL-2996_JAL-3053_tildeBracketsPipesEtcAsGapChars;hp=1326647f26e737edb5d63f7f635486e38c51b998;hpb=0afb50988743f5b902ad985793d55104e419da85;p=jalview.git

diff --git a/src/jalview/util/Comparison.java b/src/jalview/util/Comparison.java
index 1326647..aa6bcd4 100644
--- a/src/jalview/util/Comparison.java
+++ b/src/jalview/util/Comparison.java
@@ -34,14 +34,35 @@ public class Comparison
 
   private static final int TO_UPPER_CASE = 'a' - 'A';
 
-  private static final char GAP_SPACE = ' ';
+  public static final char GAP_SPACE = ' ';
 
-  private static final char GAP_DOT = '.';
+  public static final char GAP_DOT = '.';
 
-  private static final char GAP_DASH = '-';
+  public static final char GAP_DASH = '-';
 
-  public static final String GapChars = new String(new char[] { GAP_SPACE,
-      GAP_DOT, GAP_DASH });
+  public static final char GAP_TILDE = '~';
+
+  public static final char GAP_PIPE = '|';
+
+  public static final char GAP_COLON = ':';
+
+  public static final char GAP_LPAREN = '(';
+
+  public static final char GAP_RPAREN = ')';
+
+  public static final char GAP_LSQBR = '[';
+
+  public static final char GAP_RSQBR = ']';
+
+  public static final char GAP_LBRACE = '{';
+
+  public static final char GAP_RBRACE = '}';
+
+  public static final String GapChars = new String(
+          new char[]
+          { GAP_SPACE, GAP_DOT, GAP_DASH, GAP_TILDE, GAP_PIPE, GAP_COLON,
+              GAP_LPAREN,
+              GAP_RPAREN, GAP_LSQBR, GAP_RSQBR, GAP_LBRACE, GAP_RBRACE });
 
   /**
    * DOCUMENT ME!
@@ -71,7 +92,8 @@ public class Comparison
    *          int
    * @return float
    */
-  public static float compare(SequenceI ii, SequenceI jj, int start, int end)
+  public static float compare(SequenceI ii, SequenceI jj, int start,
+          int end)
   {
     String si = ii.getSequenceAsString();
     String sj = jj.getSequenceAsString();
@@ -97,8 +119,8 @@ public class Comparison
     {
       for (int j = 0; j < jlen; j++)
       {
-        if (si.substring(start + j, start + j + 1).equals(
-                sj.substring(start + j, start + j + 1)))
+        if (si.substring(start + j, start + j + 1)
+                .equals(sj.substring(start + j, start + j + 1)))
         {
           match++;
         }
@@ -112,8 +134,8 @@ public class Comparison
     {
       for (int j = 0; j < jlen; j++)
       {
-        if (si.substring(start + j, start + j + 1).equals(
-                sj.substring(start + j, start + j + 1)))
+        if (si.substring(start + j, start + j + 1)
+                .equals(sj.substring(start + j, start + j + 1)))
         {
           match++;
         }
@@ -135,7 +157,9 @@ public class Comparison
    * @param s2
    *          SequenceI
    * @return float
+   * @deprecated use PIDModel.computePID()
    */
+  @Deprecated
   public final static float PID(String seq1, String seq2)
   {
     return PID(seq1, seq2, 0, seq1.length());
@@ -144,7 +168,12 @@ public class Comparison
   static final int caseShift = 'a' - 'A';
 
   // Another pid with region specification
-  public final static float PID(String seq1, String seq2, int start, int end)
+  /**
+   * @deprecated use PIDModel.computePID()
+   */
+  @Deprecated
+  public final static float PID(String seq1, String seq2, int start,
+          int end)
   {
     return PID(seq1, seq2, start, end, true, false);
   }
@@ -165,7 +194,9 @@ public class Comparison
    * @param ungappedOnly
    *          - if true - only count PID over ungapped columns
    * @return
+   * @deprecated use PIDModel.computePID()
    */
+  @Deprecated
   public final static float PID(String seq1, String seq2, int start,
           int end, boolean wcGaps, boolean ungappedOnly)
   {
@@ -245,7 +276,24 @@ public class Comparison
    */
   public static final boolean isGap(char c)
   {
-    return (c == GAP_DASH || c == GAP_DOT || c == GAP_SPACE) ? true : false;
+    switch (c)
+    {
+    case GAP_SPACE:
+    case GAP_DOT:
+    case GAP_DASH:
+    case GAP_TILDE:
+    case GAP_PIPE:
+    case GAP_COLON:
+    case GAP_LPAREN:
+    case GAP_RPAREN:
+    case GAP_LSQBR:
+    case GAP_RSQBR:
+    case GAP_LBRACE:
+    case GAP_RBRACE:
+      return true;
+    default:
+      return false;
+    }
   }
 
   /**
@@ -274,35 +322,10 @@ public class Comparison
     {
       return false;
     }
-    char[][] letters = new char[seqs.length][];
-    for (int i = 0; i < seqs.length; i++)
-    {
-      if (seqs[i] != null)
-      {
-        char[] sequence = seqs[i].getSequence();
-        if (sequence != null)
-        {
-          letters[i] = sequence;
-        }
-      }
-    }
-
-    return areNucleotide(letters);
-  }
 
-  /**
-   * Answers true if more than 85% of the sequence residues (ignoring gaps) are
-   * A, G, C, T or U, else false. This is just a heuristic guess and may give a
-   * wrong answer (as AGCT are also amino acid codes).
-   * 
-   * @param letters
-   * @return
-   */
-  static final boolean areNucleotide(char[][] letters)
-  {
     int ntCount = 0;
     int aaCount = 0;
-    for (char[] seq : letters)
+    for (SequenceI seq : seqs)
     {
       if (seq == null)
       {
@@ -310,8 +333,10 @@ public class Comparison
       }
       // TODO could possibly make an informed guess just from the first sequence
       // to save a lengthy calculation
-      for (char c : seq)
+      int len = seq.getLength();
+      for (int i = 0; i < len; i++)
       {
+        char c = seq.getCharAt(i);
         if (isNucleotide(c))
         {
           ntCount++;
@@ -403,7 +428,7 @@ public class Comparison
     {
       return false;
     }
-    List<SequenceI> flattened = new ArrayList<SequenceI>();
+    List<SequenceI> flattened = new ArrayList<>();
     for (SequenceI[] ss : seqs)
     {
       for (SequenceI s : ss)
@@ -411,8 +436,8 @@ public class Comparison
         flattened.add(s);
       }
     }
-    final SequenceI[] oneDArray = flattened.toArray(new SequenceI[flattened
-            .size()]);
+    final SequenceI[] oneDArray = flattened
+            .toArray(new SequenceI[flattened.size()]);
     return isNucleotide(oneDArray);
   }