From a7591fdadf3edeb556d1417967787a5bfdb1da32 Mon Sep 17 00:00:00 2001
From: Jim Procter <jprocter@issues.jalview.org>
Date: Sat, 1 Oct 2016 13:00:34 +0100
Subject: [PATCH] JAL-2210 refactor code to remove from a list of dbrefs those
 which match primary dbrefs for sequences in the dataset

---
 src/jalview/analysis/CrossRef.java |   60 +++++++++++++++++++++++-------------
 1 file changed, 38 insertions(+), 22 deletions(-)

diff --git a/src/jalview/analysis/CrossRef.java b/src/jalview/analysis/CrossRef.java
index 4a7a41b..05814c2 100644
--- a/src/jalview/analysis/CrossRef.java
+++ b/src/jalview/analysis/CrossRef.java
@@ -401,28 +401,7 @@ public class CrossRef
     // first filter in case we are retrieving crossrefs that have already been
     // retrieved. this happens for cases where a database record doesn't yield
     // protein products for CDS
-    DBRefEntry[] dbrSourceSet = sourceRefs.toArray(new DBRefEntry[0]);
-    for (SequenceI sq : dataset.getSequences())
-    {
-      boolean dupeFound = false;
-      // !fromDna means we are looking only for nucleotide sequences, not
-      // protein
-      if (sq.isProtein() == fromDna)
-      {
-        for (DBRefEntry dbr : sq.getPrimaryDBRefs())
-        {
-          for (DBRefEntry found : DBRefUtils.searchRefs(dbrSourceSet, dbr))
-          {
-            sourceRefs.remove(found);
-            dupeFound = true;
-          }
-        }
-      }
-      if (dupeFound)
-      {
-        dbrSourceSet = sourceRefs.toArray(new DBRefEntry[0]);
-      }
-    }
+    removeAlreadyRetrievedSeqs(sourceRefs, fromDna);
     if (sourceRefs.size() == 0)
     {
       // no more work to do! We already had all requested sequence records in
@@ -451,6 +430,43 @@ public class CrossRef
                 : retrievedSequence.getDatasetSequence();
         DBRefEntry[] dbr = retrievedSequence.getDBRefs();
         if (dbr != null)
+
+  /**
+   * Search dataset for sequences with a primary reference contained in
+   * sourceRefs.
+   * 
+   * @param sourceRefs
+   *          - list of references to filter.
+   * @param fromDna
+   *          - type of sequence to search for matching primary reference.
+   */
+  private void removeAlreadyRetrievedSeqs(List<DBRefEntry> sourceRefs,
+          boolean fromDna)
+  {
+    DBRefEntry[] dbrSourceSet = sourceRefs.toArray(new DBRefEntry[0]);
+    for (SequenceI sq : dataset.getSequences())
+    {
+      boolean dupeFound = false;
+      // !fromDna means we are looking only for nucleotide sequences, not
+      // protein
+      if (sq.isProtein() == fromDna)
+      {
+        for (DBRefEntry dbr : sq.getPrimaryDBRefs())
+        {
+          for (DBRefEntry found : DBRefUtils.searchRefs(dbrSourceSet, dbr))
+          {
+            sourceRefs.remove(found);
+            dupeFound = true;
+          }
+        }
+      }
+      if (dupeFound)
+      {
+        dbrSourceSet = sourceRefs.toArray(new DBRefEntry[0]);
+      }
+    }
+  }
+
         {
           for (DBRefEntry dbref : dbr)
           {
-- 
1.7.10.2