JAL-2154 Optional message parameter for assertDatasetNormalised

[jalview.git] / test / jalview / datamodel / AlignmentTest.java

diff --git a/test/jalview/datamodel/AlignmentTest.java b/test/jalview/datamodel/AlignmentTest.java
index 379a9c5..5568c8c 100644 (file)
--- a/test/jalview/datamodel/AlignmentTest.java
+++ b/test/jalview/datamodel/AlignmentTest.java
@@ -419,7 +419,133 @@ public class AlignmentTest
     al.getDataset().addSequence(sqanotherout);
     assertVerifyAlignment(al, false,
             "verify should have failed when a sequence was added twice to the dataset");
+    al.getDataset().deleteSequence(sqanotherout);
+    assertVerifyAlignment(al, true,
+            "verify should have passed after duplicate entry for sequence was removed");
+  }
+
+  /**
+   * checks that the sequence data for an alignment's dataset is non-redundant.
+   * Fails if there are sequences with same id, sequence, start, and.
+   */
+
+  public static void assertDatasetIsNormalised(AlignmentI al)
+  {
+    assertAlignmentDatasetRefs(al, null);
+  }
+  public static void assertDatasetIsNormalised(AlignmentI al, String message)
+  {
+    if (al.getDataset()!=null)
+    {
+      assertDatasetIsNormalised(al.getDataset(), message);
+      return;
+    }
+    /*
+     * look for pairs of sequences with same ID, start, end, and sequence
+     */
+    List<SequenceI> seqSet = al.getSequences();
+    for (int p=0;p<seqSet.size(); p++)
+    {
+      SequenceI pSeq = seqSet.get(p);
+      for (int q=p+1; q<seqSet.size(); q++)
+      {
+        SequenceI qSeq = seqSet.get(q);
+        if (pSeq.getStart()!=qSeq.getStart())
+        {
+          continue;
+        }
+        if (pSeq.getEnd()!=qSeq.getEnd())
+        {
+          continue;
+        }
+        if (!pSeq.getName().equals(qSeq.getName()))
+        {
+          continue;
+        }
+        if (!Arrays.equals(pSeq.getSequence(), qSeq.getSequence()))
+        {
+          continue;
+        }
+        Assert.fail((message == null ? "" : message + " :")
+                + "Found similar sequences at position " + p + " and " + q
+                + "\n" + pSeq.toString());
+      }
+    }
+  }
+  
+  @Test(groups = { "Functional", "Asserts" })
+  public void testAssertDatasetIsNormalised()
+  {
+    Sequence sq1 = new Sequence("s1/1-4", "asdf");
+    Sequence sq1shift = new Sequence("s1/2-5", "asdf");
+    Sequence sq1seqd = new Sequence("s1/1-4", "asdt");
+    Sequence sq2 = new Sequence("s2/1-4", "asdf");
+    Sequence sq1dup = new Sequence("s1/1-4", "asdf");
+
+    Alignment al = new Alignment(new SequenceI[] { sq1 });
+    al.setDataset(null);
+
+    try
+    {
+      assertDatasetIsNormalised(al);
+    } catch (AssertionError ae)
+    {
+      Assert.fail("Single sequence should be valid normalised dataset.");
+    }
+    al.addSequence(sq2);
+    try
+    {
+      assertDatasetIsNormalised(al);
+    } catch (AssertionError ae)
+    {
+      Assert.fail("Two different sequences should be valid normalised dataset.");
+    }
+    /*
+     * now change sq2's name in the alignment. should still be valid
+     */
+    al.findName(sq2.getName()).setName("sq1");
+    try
+    {
+      assertDatasetIsNormalised(al);
+    } catch (AssertionError ae)
+    {
+      Assert.fail("Two different sequences in dataset, but same name in alignment, should be valid normalised dataset.");
+    }
+
+    al.addSequence(sq1seqd);
+    try
+    {
+      assertDatasetIsNormalised(al);
+    } catch (AssertionError ae)
+    {
+      Assert.fail("sq1 and sq1 with different sequence should be distinct.");
+    }
+
+    al.addSequence(sq1shift);
+    try
+    {
+      assertDatasetIsNormalised(al);
+    } catch (AssertionError ae)
+    {
+      Assert.fail("sq1 and sq1 with different start/end should be distinct.");
+    }
+    /*
+     * finally, the failure case
+     */
+    al.addSequence(sq1dup);
+    boolean ssertRaised = false;
+    try
+    {
+      assertDatasetIsNormalised(al);
 
+    } catch (AssertionError ae)
+    {
+      ssertRaised = true;
+    }
+    if (!ssertRaised)
+    {
+      Assert.fail("Expected identical sequence to raise exception.");
+    }
   }
   /*
    * Read in Stockholm format test data including secondary structure
@@ -857,6 +983,27 @@ public class AlignmentTest
   }
 
   @Test(groups = "Functional")
+  public void testAddSequencePreserveDatasetIntegrity()
+  {
+    Sequence seq = new Sequence("testSeq", "ABCDEFGHIJKLMNOPQRSTUVWXYZ");
+    Alignment align = new Alignment(new SequenceI[] { seq });
+    align.createDatasetAlignment();
+    AlignmentI ds = align.getDataset();
+    SequenceI copy = new Sequence(seq);
+    copy.insertCharAt(3, 5, '-');
+    align.addSequence(copy);
+    Assert.assertEquals(align.getDataset().getHeight(), 1,
+            "Dataset shouldn't have more than one sequence.");
+
+    Sequence seq2 = new Sequence("newtestSeq", "ABCDEFGHIJKLMNOPQRSTUVWXYZ");
+    align.addSequence(seq2);
+    Assert.assertEquals(align.getDataset().getHeight(), 2,
+            "Dataset should now have two sequences.");
+
+    assertAlignmentDatasetRefs(align,
+            "addSequence broke dataset reference integrity");
+  }
+  @Test(groups = "Functional")
   public void getVisibleStartAndEndIndexTest()
   {
     Sequence seq = new Sequence("testSeq", "ABCDEFGHIJKLMNOPQRSTUVWXYZ");
@@ -884,4 +1031,5 @@ public class AlignmentTest
     assertEquals(1, startEnd[0]);
     assertEquals(23, startEnd[1]);
   }
+
 }