GATK read group requirement iron hand

-- The GATK will now throw a user exception if it opens a SAM/BAM file that doesn't have at least one RG defined
-- LIBS again throws an error if the complete list of samples isn't provided
-- Updating ExmpleCountLociPipeline test to use the well-formated versions of the exampleBAM and exampleFASTA files in testdata, instead of the old broken ones in validation_data.
-- Convenience constructors for UserExceptions.MalformedBAM

public/java/src/org/broadinstitute/sting/gatk/datasources/reads/SAMDataSource.java

@@ -235,6 +235,12 @@ public class SAMDataSource {
         for(SAMFileReader reader: readers.values()) {
             // Get the sort order, forcing it to coordinate if unsorted.
             SAMFileHeader header = reader.getFileHeader();
+
+            if ( header.getReadGroups().isEmpty() ) {
+                throw new UserException.MalformedBAM(readers.getReaderID(reader).samFile,
+                        "SAM file doesn't have any read groups defined in the header.  The GATK no longer supports SAM files without read groups");
+            }
+
             SAMFileHeader.SortOrder sortOrder = header.getSortOrder() != SAMFileHeader.SortOrder.unsorted ? header.getSortOrder() : SAMFileHeader.SortOrder.coordinate;
 
             // Validate that all input files are sorted in the same order.

public/java/src/org/broadinstitute/sting/gatk/iterators/LocusIteratorByState.java

@@ -283,10 +283,7 @@ public class LocusIteratorByState extends LocusIterator {
         // currently the GATK expects this LocusIteratorByState to accept empty sample lists, when
         // there's no read data.  So we need to throw this error only when samIterator.hasNext() is true
         if ( this.samples.isEmpty() && samIterator.hasNext() ) {
-            // actually we cannot process BAMs without read groups unless we tolerate empty
-            // sample lists.  In the empty case we need to add the null element to the samples
-            this.samples.add(null);
-            //throw new IllegalArgumentException("samples list must not be empty");
+            throw new IllegalArgumentException("samples list must not be empty");
         }
     }
 

public/java/src/org/broadinstitute/sting/utils/exceptions/UserException.java

@@ -160,7 +160,15 @@ public class UserException extends ReviewedStingException {
 
     public static class MalformedBAM extends UserException {
         public MalformedBAM(SAMRecord read, String message) {
-            super(String.format("SAM/BAM file %s is malformed: %s", read.getFileSource() != null ? read.getFileSource().getReader() : "(none)", message));
+            this(read.getFileSource() != null ? read.getFileSource().getReader().toString() : "(none)", message);
+        }
+
+        public MalformedBAM(File file, String message) {
+            this(file.toString(), message);
+        }
+
+        public MalformedBAM(String source, String message) {
+            super(String.format("SAM/BAM file %s is malformed: %s", source, message));
         }
     }
 

public/scala/test/org/broadinstitute/sting/queue/pipeline/examples/ExampleCountLociPipelineTest.scala

@@ -36,8 +36,8 @@ class ExampleCountLociPipelineTest {
     spec.name = "countloci"
     spec.args = Array(
       " -S public/scala/qscript/org/broadinstitute/sting/queue/qscripts/examples/ExampleCountLoci.scala",
-      " -R " + BaseTest.hg18Reference,
-      " -I " + BaseTest.validationDataLocation + "small_bam_for_countloci.bam",
+      " -R " + BaseTest.testDir + "exampleFASTA.fasta",
+      " -I " + BaseTest.testDir + "exampleBAM.bam",
       " -o " + testOut).mkString
     spec.fileMD5s += testOut -> "67823e4722495eb10a5e4c42c267b3a6"
     PipelineTest.executeTest(spec)