X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fio%2FStockholmFile.java;h=ccbe8180477364cb52db5158b88de89af00df4c2;hb=f683f4d6c8af43be0ffeb96513b52572619efed5;hp=dedbd440d9ac5c31d17f25f3ff06f5665a0acc2a;hpb=0812da43d6b8d673aa25d2d7a7c5f65aab1b1642;p=jalview.git

diff --git a/src/jalview/io/StockholmFile.java b/src/jalview/io/StockholmFile.java
index dedbd44..ccbe818 100644
--- a/src/jalview/io/StockholmFile.java
+++ b/src/jalview/io/StockholmFile.java
@@ -31,9 +31,15 @@ import jalview.datamodel.*;
 
 /**
  * This class is supposed to parse a Stockholm format file into Jalview
- * 
+ * There are TODOs in this class: we do not know what the database source and
+ * version is for the file when parsing the #GS= AC tag which associates accessions
+ * with sequences. 
+ * Database references are also not parsed correctly: a separate reference string
+ * parser must be added to parse the database reference form into Jalview's local
+ * representation.
  * @author bsb at sanger.ac.uk
- * @version 0.3
+ * @version 0.3 + jalview mods
+ * 
  */
 public class StockholmFile extends AlignFile
 {
@@ -47,7 +53,10 @@ public class StockholmFile extends AlignFile
   {
     super(inFile, type);
   }
-
+  public StockholmFile(FileParse source) throws IOException
+  {
+    super(source);
+  }
   public void initData()
   {
     super.initData();
@@ -91,7 +100,7 @@ public class StockholmFile extends AlignFile
     rend = new Regex("\\/\\/"); // Find the end of an alignment
     p = new Regex("(\\S+)\\/(\\d+)\\-(\\d+)"); // split sequence id in
                                                 // id/from/to
-    s = new Regex("(\\S+)\\s+(\\w{2})\\s+(.*)"); // Parses annotation subtype
+    s = new Regex("(\\S+)\\s+(\\S*)\\s+(.*)"); // Parses annotation subtype
     r = new Regex("#=(G[FSRC]?)\\s+(.*)"); // Finds any annotation line
     x = new Regex("(\\S+)\\s+(\\S+)"); // split id from sequence
 
@@ -118,7 +127,7 @@ public class StockholmFile extends AlignFile
         {
           String acc = (String) accs.nextElement();
           // logger.debug("Processing sequence " + acc);
-          String seq = (String) seqs.get(acc);
+          String seq = (String) seqs.remove(acc);
           if (maxLength < seq.length())
           {
             maxLength = seq.length();
@@ -131,7 +140,7 @@ public class StockholmFile extends AlignFile
 
           if (seqAnn != null && seqAnn.containsKey(acc))
           {
-            accAnnotations = (Hashtable) seqAnn.get(acc);
+            accAnnotations = (Hashtable) seqAnn.remove(acc);
           }
 
           // Split accession in id and from/to
@@ -158,16 +167,15 @@ public class StockholmFile extends AlignFile
             {
               String src = dbr.substring(0, dbr.indexOf(";"));
               String acn = dbr.substring(dbr.indexOf(";") + 1);
-              DBRefEntry dbref = new DBRefEntry(jalview.util.DBRefUtils
-                      .getCanonicalName(src), acn, "");
-              seqO.addDBRef(dbref);
+              jalview.util.DBRefUtils.parseToDbRef(seqO, src, "0", acn);
+              //seqO.addDBRef(dbref);
             }
           }
           Hashtable features = null;
           // We need to adjust the positions of all features to account for gaps
           try
           {
-            features = (Hashtable) accAnnotations.get("features");
+            features = (Hashtable) accAnnotations.remove("features");
           } catch (java.lang.NullPointerException e)
           {
             // loggerwarn("Getting Features for " + acc + ": " +
@@ -177,6 +185,7 @@ public class StockholmFile extends AlignFile
           // if we have features
           if (features != null)
           {
+            int posmap[] = seqO.findPositionMap();
             Enumeration i = features.keys();
             while (i.hasMoreElements())
             {
@@ -185,8 +194,7 @@ public class StockholmFile extends AlignFile
               // TODO: parse out scores as annotation row
               // TODO: map coding region to core jalview feature types
               String type = i.nextElement().toString();
-              Hashtable content = (Hashtable) features.get(type);
-
+              Hashtable content = (Hashtable) features.remove(type);
               Enumeration j = content.keys();
               while (j.hasMoreElements())
               {
@@ -196,9 +204,9 @@ public class StockholmFile extends AlignFile
                 for (int k = 0; k < byChar.length; k++)
                 {
                   char c = byChar[k];
-                  if (!(c == ' ' || c == '_' || c == '-'))
+                  if (!(c == ' ' || c == '_' || c == '-' || c == '.')) // PFAM uses '.' for feature background
                   {
-                    int new_pos = seqO.findPosition(k);
+                    int new_pos = posmap[k]; // look up nearest seqeunce position to this column
                     SequenceFeature feat = new SequenceFeature(type, desc,
                             new_pos, new_pos, 0f, null);
 
@@ -208,12 +216,15 @@ public class StockholmFile extends AlignFile
               }
 
             }
-
+            
           }
+          // garbage collect
+          
           // logger.debug("Adding seq " + acc + " from " + start + " to " + end
           // + ": " + seq);
           this.seqs.addElement(seqO);
         }
+        return; // finished parsing this segment of source
       }
       else if (!r.search(line))
       {
@@ -365,14 +376,18 @@ public class StockholmFile extends AlignFile
           {
             String acc = s.stringMatched(1);
             String type = s.stringMatched(2);
-            String seq = s.stringMatched(3);
-            String description = new String();
-
+            String seq = new String(s.stringMatched(3));
+            String description = null;
             // Check for additional information about the current annotation
-            if (x.search(seq))
+            // We use a simple  string tokenizer here for speed
+            StringTokenizer sep = new StringTokenizer(seq," \t");
+            description = sep.nextToken();
+            if (sep.hasMoreTokens())
             {
-              description = x.stringMatched(1);
-              seq = x.stringMatched(2);
+              seq = sep.nextToken();
+            } else {
+              seq = description;
+              description = new String();
             }
             // sequence id with from-to fields
 
@@ -427,7 +442,8 @@ public class StockholmFile extends AlignFile
           }
           else
           {
-            throw new IOException("Error parsing " + line);
+            System.err.println("Warning - couldn't parse sequence annotation row line:\n"+line);
+            // throw new IOException("Error parsing " + line);
           }
         }
         else
@@ -464,18 +480,18 @@ public class StockholmFile extends AlignFile
     {
       String pos = annots.substring(i, i + 1);
       Annotation ann;
-      ann = new Annotation(pos, "", ' ', Float.NaN);
+      ann = new Annotation(pos, "", ' ', 0f); // 0f is 'valid' null - will not be written out 
       if (ss)
       {
         ann.secondaryStructure = jalview.schemes.ResidueProperties
                 .getDssp3state(pos).charAt(0);
         if (ann.secondaryStructure == pos.charAt(0) || pos.charAt(0) == 'C')
         {
-          ann.displayCharacter = "";
+          ann.displayCharacter = ""; // null; // " ";
         }
         else
         {
-          ann.displayCharacter += " ";
+          ann.displayCharacter = " "+ann.displayCharacter;
         }
       }
 
@@ -493,7 +509,7 @@ public class StockholmFile extends AlignFile
     if (annot == null)
     {
       annot = new AlignmentAnnotation(type, type, els);
-      annotation.add(annot);
+      annotation.addElement(annot);
     }
     else
     {