Add support for block gzipped files with a .bgz suffix (as well as

a .gz suffix), see HadoopGenomics/Hadoop-BAM#106. Also remove spurious debug.
broadinstitute · Jul 1, 2016 · a4c8e5b · a4c8e5b
1 parent 3f08383
commit a4c8e5b
Show file tree

Hide file tree

Showing 2 changed files with 3 additions and 7 deletions.
diff --git a/...main/java/org/broadinstitute/hellbender/engine/spark/datasources/VariantsSparkSource.java b/...main/java/org/broadinstitute/hellbender/engine/spark/datasources/VariantsSparkSource.java
@@ -13,6 +13,7 @@
 import org.seqdoop.hadoop_bam.VCFInputFormat;
 import org.seqdoop.hadoop_bam.VariantContextWritable;
 import org.seqdoop.hadoop_bam.util.BGZFCodec;
+import org.seqdoop.hadoop_bam.util.BGZFEnhancedGzipCodec;
 
 import java.util.List;
 
@@ -61,7 +62,8 @@ public JavaRDD<GATKVariant> getParallelVariants(final List<String> vcfs, final L
      */
     public JavaRDD<VariantContext> getParallelVariantContexts(final String vcf, final List<SimpleInterval> intervals) {
         Configuration conf = new Configuration();
-        conf.set("io.compression.codecs", BGZFCodec.class.getCanonicalName());
+        conf.setStrings("io.compression.codecs", BGZFEnhancedGzipCodec.class.getCanonicalName(),
+                BGZFCodec.class.getCanonicalName());
         if (intervals != null && !intervals.isEmpty()) {
             VCFInputFormat.setIntervals(conf, intervals);
         }

diff --git a/...rg/broadinstitute/hellbender/tools/spark/pipelines/CountVariantsSparkIntegrationTest.java b/...rg/broadinstitute/hellbender/tools/spark/pipelines/CountVariantsSparkIntegrationTest.java
@@ -1,16 +1,13 @@
 package org.broadinstitute.hellbender.tools.spark.pipelines;
 
-import htsjdk.samtools.util.BlockCompressedInputStream;
 import org.apache.commons.io.FileUtils;
 import org.broadinstitute.hellbender.CommandLineProgramTest;
 import org.broadinstitute.hellbender.utils.test.ArgumentsBuilder;
 import org.testng.Assert;
 import org.testng.annotations.DataProvider;
 import org.testng.annotations.Test;
 
-import java.io.BufferedInputStream;
 import java.io.File;
-import java.io.FileInputStream;
 
 public final class CountVariantsSparkIntegrationTest extends CommandLineProgramTest {
 
@@ -64,9 +61,6 @@ public Object[][] intervals(){
 
     @Test(dataProvider = "intervals", groups = "spark")
     public void testCountVariantsWithIntervals(final File fileIn, final String intervalArgs, final long expected) throws Exception {
-
-        File f = new File(largeFileTestDir, "dbsnp_138.b37.20.21.vcf.blockgz.gz");
-        System.out.println("tw: " + BlockCompressedInputStream.isValidFile(new BufferedInputStream(new FileInputStream(f))));
         final File outputTxt = createTempFile("count_variants", ".txt");
         ArgumentsBuilder args = new ArgumentsBuilder();
         args.addVCF(fileIn);