Merge branch 'feature/JAL-3692enaEndpoint' into patch/for2.11.2/JAL-3821_ena_rna_moltype

[jalview.git] / src / jalview / ws / dbsources / EmblXmlSource.java

diff --git a/src/jalview/ws/dbsources/EmblXmlSource.java b/src/jalview/ws/dbsources/EmblXmlSource.java
index 19366e0..97d7c9f 100644 (file)
--- a/src/jalview/ws/dbsources/EmblXmlSource.java
+++ b/src/jalview/ws/dbsources/EmblXmlSource.java
@@ -20,6 +20,27 @@
  */
 package jalview.ws.dbsources;
 
+import java.io.File;
+import java.io.FileInputStream;
+import java.io.InputStream;
+import java.text.ParseException;
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Hashtable;
+import java.util.List;
+import java.util.Map;
+import java.util.Map.Entry;
+
+import javax.xml.bind.JAXBContext;
+import javax.xml.bind.JAXBElement;
+import javax.xml.bind.JAXBException;
+import javax.xml.stream.FactoryConfigurationError;
+import javax.xml.stream.XMLInputFactory;
+import javax.xml.stream.XMLStreamException;
+import javax.xml.stream.XMLStreamReader;
+
+import com.stevesoft.pat.Regex;
+
 import jalview.analysis.SequenceIdMatcher;
 import jalview.bin.Cache;
 import jalview.datamodel.Alignment;
@@ -35,36 +56,23 @@ import jalview.util.DBRefUtils;
 import jalview.util.DnaUtils;
 import jalview.util.MapList;
 import jalview.util.MappingUtils;
-import jalview.util.MessageManager;
 import jalview.ws.ebi.EBIFetchClient;
 import jalview.xml.binding.embl.EntryType;
 import jalview.xml.binding.embl.EntryType.Feature;
 import jalview.xml.binding.embl.EntryType.Feature.Qualifier;
-import jalview.xml.binding.jalview.JalviewModel;
 import jalview.xml.binding.embl.ROOT;
 import jalview.xml.binding.embl.XrefType;
 
-import java.io.File;
-import java.io.FileInputStream;
-import java.io.InputStream;
-import java.text.ParseException;
-import java.util.ArrayList;
-import java.util.Arrays;
-import java.util.Hashtable;
-import java.util.List;
-import java.util.Map;
-import java.util.Map.Entry;
-
-import javax.xml.bind.JAXBContext;
-import javax.xml.bind.JAXBElement;
-import javax.xml.bind.JAXBException;
-import javax.xml.stream.FactoryConfigurationError;
-import javax.xml.stream.XMLInputFactory;
-import javax.xml.stream.XMLStreamException;
-import javax.xml.stream.XMLStreamReader;
-
+/**
+ * Provides XML binding and parsing of EMBL or EMBLCDS records retrieved from
+ * (e.g.) {@code https://www.ebi.ac.uk/ena/data/view/x53828&display=xml}.
+ * 
+ * @deprecated endpoint withdrawn August 2020 (JAL-3692), use EmblFlatfileSource
+ */
 public abstract class EmblXmlSource extends EbiFileRetrievedProxy
 {
+  private static final Regex ACCESSION_REGEX = new Regex("^[A-Z]+[0-9]+");
+
   /*
    * JAL-1856 Embl returns this text for query not found
    */
@@ -99,9 +107,10 @@ public abstract class EmblXmlSource extends EbiFileRetrievedProxy
     } catch (Exception e)
     {
       stopQuery();
-      throw new Exception(MessageManager.formatMessage(
-              "exception.ebiembl_retrieval_failed_on", new String[]
-              { emprefx.toLowerCase(), query.trim() }), e);
+      throw new Exception(
+              String.format("EBI EMBL XML retrieval failed for %s:%s",
+                      emprefx.toLowerCase(), query.trim()),
+              e);
     }
     return getEmblSequenceRecords(emprefx, query, reply);
   }
@@ -183,7 +192,8 @@ public abstract class EmblXmlSource extends EbiFileRetrievedProxy
       XMLStreamReader streamReader = XMLInputFactory.newInstance()
               .createXMLStreamReader(is);
       javax.xml.bind.Unmarshaller um = jc.createUnmarshaller();
-      JAXBElement<ROOT> rootElement =  um.unmarshal(streamReader, ROOT.class);
+      JAXBElement<ROOT> rootElement = um.unmarshal(streamReader,
+              ROOT.class);
       ROOT root = rootElement.getValue();
 
       /*
@@ -564,6 +574,7 @@ public abstract class EmblXmlSource extends EbiFileRetrievedProxy
               proteinSeq = new Sequence(proteinSeqName,
                       product.getSequenceAsString());
               matcher.add(proteinSeq);
+              proteinSeq.setDescription(product.getDescription());
               peptides.add(proteinSeq);
             }
             dnaToProteinMapping.setTo(proteinSeq);
@@ -617,8 +628,7 @@ public abstract class EmblXmlSource extends EbiFileRetrievedProxy
               && dnaToProteinMapping.getTo() != null)
       {
         DBRefEntry dnaToEmblProteinRef = new DBRefEntry(
-                DBRefSource.EMBLCDSProduct, sequenceVersion,
-                proteinId);
+                DBRefSource.EMBLCDSProduct, sequenceVersion, proteinId);
         dnaToEmblProteinRef.setMap(dnaToProteinMapping);
         dnaToProteinMapping.setMappedFromId(proteinId);
         dna.addDBRef(dnaToEmblProteinRef);
@@ -647,7 +657,7 @@ public abstract class EmblXmlSource extends EbiFileRetrievedProxy
     {
       return new int[] {};
     }
-  
+
     try
     {
       List<int[]> ranges = DnaUtils.parseLocation(location);
@@ -711,6 +721,40 @@ public abstract class EmblXmlSource extends EbiFileRetrievedProxy
     return sf;
   }
 
+  @Override
+  public String getAccessionSeparator()
+  {
+    return null;
+  }
+
+  @Override
+  public Regex getAccessionValidator()
+  {
+    return ACCESSION_REGEX;
+  }
+
+  @Override
+  public String getDbVersion()
+  {
+    return "0";
+  }
+
+  @Override
+  public int getTier()
+  {
+    return 0;
+  }
+
+  @Override
+  public boolean isValidReference(String accession)
+  {
+    if (accession == null || accession.length() < 2)
+    {
+      return false;
+    }
+    return getAccessionValidator().search(accession);
+  }
+
   /**
    * Truncates (if necessary) the exon intervals to match 3 times the length of
    * the protein; also accepts 3 bases longer (for stop codon not included in
@@ -729,7 +773,7 @@ public abstract class EmblXmlSource extends EbiFileRetrievedProxy
     }
     int expectedCdsLength = proteinLength * 3;
     int exonLength = MappingUtils.getLength(Arrays.asList(exon));
-  
+
     /*
      * if exon length matches protein, or is shorter, or longer by the 
      * length of a stop codon (3 bases), then leave it unchanged
@@ -739,7 +783,7 @@ public abstract class EmblXmlSource extends EbiFileRetrievedProxy
     {
       return exon;
     }
-  
+
     int origxon[];
     int sxpos = -1;
     int endxon = 0;
@@ -759,7 +803,7 @@ public abstract class EmblXmlSource extends EbiFileRetrievedProxy
           // .println("Truncating final exon interval on region by "
           // + (cdspos - cdslength));
         }
-  
+
         /*
          * shrink the final exon - reduce end position if forward
          * strand, increase it if reverse
@@ -775,7 +819,7 @@ public abstract class EmblXmlSource extends EbiFileRetrievedProxy
         break;
       }
     }
-  
+
     if (sxpos != -1)
     {
       // and trim the exon interval set if necessary