Merge commit 'alpha/update_2_12_for_2_11_2_series_merge^2' into HEAD

[jalview.git] / src / jalview / analysis / AlignmentUtils.java

diff --git a/src/jalview/analysis/AlignmentUtils.java b/src/jalview/analysis/AlignmentUtils.java
index 4cd1fc4..7c1f4d7 100644 (file)
--- a/src/jalview/analysis/AlignmentUtils.java
+++ b/src/jalview/analysis/AlignmentUtils.java
@@ -20,6 +20,8 @@
  */
 package jalview.analysis;
 
+import java.util.Locale;
+
 import jalview.commands.RemoveGapColCommand;
 import jalview.datamodel.AlignedCodon;
 import jalview.datamodel.AlignedCodonFrame;
@@ -181,9 +183,9 @@ public class AlignmentUtils
       // TODO use Character.toLowerCase to avoid creating String objects?
       char[] upstream = new String(ds
               .getSequence(s.getStart() - 1 - ustream_ds, s.getStart() - 1))
-                      .toLowerCase().toCharArray();
+                      .toLowerCase(Locale.ROOT).toCharArray();
       char[] downstream = new String(
-              ds.getSequence(s_end - 1, s_end + dstream_ds)).toLowerCase()
+              ds.getSequence(s_end - 1, s_end + dstream_ds)).toLowerCase(Locale.ROOT)
                       .toCharArray();
       char[] coreseq = s.getSequence();
       char[] nseq = new char[offset + upstream.length + downstream.length
@@ -462,7 +464,7 @@ public class AlignmentUtils
     if (cdnaLength != mappedLength && cdnaLength > 2)
     {
       String lastCodon = String.valueOf(cdnaSeqChars,
-              cdnaLength - CODON_LENGTH, CODON_LENGTH).toUpperCase();
+              cdnaLength - CODON_LENGTH, CODON_LENGTH).toUpperCase(Locale.ROOT);
       for (String stop : ResidueProperties.STOP_CODONS)
       {
         if (lastCodon.equals(stop))
@@ -479,7 +481,7 @@ public class AlignmentUtils
      */
     int startOffset = 0;
     if (cdnaLength != mappedLength && cdnaLength > 2
-            && String.valueOf(cdnaSeqChars, 0, CODON_LENGTH).toUpperCase()
+            && String.valueOf(cdnaSeqChars, 0, CODON_LENGTH).toUpperCase(Locale.ROOT)
                     .equals(ResidueProperties.START))
     {
       startOffset += CODON_LENGTH;
@@ -1459,28 +1461,31 @@ public class AlignmentUtils
       final List<AlignmentAnnotation> result = new ArrayList<>();
       for (AlignmentAnnotation dsann : datasetAnnotations)
       {
-        /*
-         * Find matching annotations on the alignment. If none is found, then
-         * add this annotation to the list of 'addable' annotations for this
-         * sequence.
-         */
-        final Iterable<AlignmentAnnotation> matchedAlignmentAnnotations = al
-                .findAnnotations(seq, dsann.getCalcId(), dsann.label);
-        if (!matchedAlignmentAnnotations.iterator().hasNext())
+        if (dsann.annotations != null) // ignore non-positional annotation
         {
-          result.add(dsann);
-          if (labelForCalcId != null)
+          /*
+           * Find matching annotations on the alignment. If none is found, then
+           * add this annotation to the list of 'addable' annotations for this
+           * sequence.
+           */
+          final Iterable<AlignmentAnnotation> matchedAlignmentAnnotations = al
+                  .findAnnotations(seq, dsann.getCalcId(), dsann.label);
+          if (!matchedAlignmentAnnotations.iterator().hasNext())
           {
-            labelForCalcId.put(dsann.getCalcId(), dsann.label);
+            result.add(dsann);
+            if (labelForCalcId != null)
+            {
+              labelForCalcId.put(dsann.getCalcId(), dsann.label);
+            }
           }
         }
-      }
-      /*
-       * Save any addable annotations for this sequence
-       */
-      if (!result.isEmpty())
-      {
-        candidates.put(seq, result);
+        /*
+         * Save any addable annotations for this sequence
+         */
+        if (!result.isEmpty())
+        {
+          candidates.put(seq, result);
+        }
       }
     }
   }
@@ -1732,15 +1737,8 @@ public class AlignmentUtils
 
           cdsSeqs.add(cdsSeq);
 
-          if (!dataset.getSequences().contains(cdsSeqDss))
-          {
-            // check if this sequence is a newly created one
-            // so needs adding to the dataset
-            dataset.addSequence(cdsSeqDss);
-          }
-
           /*
-           * add a mapping from CDS to the (unchanged) mapped to range
+           * build the mapping from CDS to protein
            */
           List<int[]> cdsRange = Collections
                   .singletonList(new int[]
@@ -1749,16 +1747,26 @@ public class AlignmentUtils
           MapList cdsToProteinMap = new MapList(cdsRange,
                   mapList.getToRanges(), mapList.getFromRatio(),
                   mapList.getToRatio());
-          AlignedCodonFrame cdsToProteinMapping = new AlignedCodonFrame();
-          cdsToProteinMapping.addMap(cdsSeqDss, proteinProduct,
-                  cdsToProteinMap);
 
-          /*
-           * guard against duplicating the mapping if repeating this action
-           */
-          if (!mappings.contains(cdsToProteinMapping))
+          if (!dataset.getSequences().contains(cdsSeqDss))
           {
-            mappings.add(cdsToProteinMapping);
+            /*
+             * if this sequence is a newly created one, add it to the dataset
+             * and made a CDS to protein mapping (if sequence already exists,
+             * CDS-to-protein mapping _is_ the transcript-to-protein mapping)
+             */
+            dataset.addSequence(cdsSeqDss);
+            AlignedCodonFrame cdsToProteinMapping = new AlignedCodonFrame();
+            cdsToProteinMapping.addMap(cdsSeqDss, proteinProduct,
+                  cdsToProteinMap);
+
+            /*
+             * guard against duplicating the mapping if repeating this action
+             */
+            if (!mappings.contains(cdsToProteinMapping))
+            {
+              mappings.add(cdsToProteinMapping);
+            }
           }
 
           propagateDBRefsToCDS(cdsSeqDss, dnaSeq.getDatasetSequence(),
@@ -2584,6 +2592,13 @@ public class AlignmentUtils
     {
       List<SequenceI> alignedSequences = alignedDatasets
               .get(seq.getDatasetSequence());
+      if (alignedSequences.isEmpty())
+      {
+        /*
+         * defensive check - shouldn't happen! (JAL-3536)
+         */
+        continue;
+      }
       SequenceI alignedSeq = alignedSequences.get(0);
 
       /*