?count as refactor?

broadinstitute · Jul 11, 2018 · 97e32cb · 97e32cb
1 parent e3211fe
commit 97e32cb
Show file tree

Hide file tree

Showing 3 changed files with 36 additions and 27 deletions.
diff --git a/src/main/java/org/broadinstitute/hellbender/engine/GATKTool.java b/src/main/java/org/broadinstitute/hellbender/engine/GATKTool.java
@@ -9,7 +9,6 @@
 import htsjdk.variant.variantcontext.writer.Options;
 import htsjdk.variant.variantcontext.writer.VariantContextWriter;
 import htsjdk.variant.vcf.VCFHeaderLine;
-import htsjdk.variant.vcf.VCFSimpleHeaderLine;
 import org.broadinstitute.barclay.argparser.Argument;
 import org.broadinstitute.barclay.argparser.ArgumentCollection;
 import org.broadinstitute.barclay.argparser.CommandLinePluginDescriptor;
@@ -836,17 +835,13 @@ protected SAMFileHeader getHeaderForSAMWriter(){
      * date and command line, otherwise an empty set.
      */
     protected Set<VCFHeaderLine> getDefaultToolVCFHeaderLines() {
-        final Set<VCFHeaderLine> gatkToolHeaderLines = new HashSet<>();
         if (addOutputVCFCommandLine) {
-            final Map<String, String> simpleHeaderLineMap = new HashMap<>(4);
-            simpleHeaderLineMap.put("ID", this.getClass().getSimpleName());
-            simpleHeaderLineMap.put("Version", getVersion());
-            simpleHeaderLineMap.put("Date", Utils.getDateTimeForDisplay((ZonedDateTime.now())));
-            simpleHeaderLineMap.put("CommandLine", getCommandLine());
-            gatkToolHeaderLines.add(new VCFHeaderLine("source", this.getClass().getSimpleName()));
-            gatkToolHeaderLines.add(new VCFSimpleHeaderLine(String.format("%sCommandLine", getToolkitShortName()), simpleHeaderLineMap));
+            return GATKVariantContextUtils
+                    .getDefaultVCFHeaderLines(getToolkitShortName(), this.getClass().getSimpleName(),
+                            getVersion(), Utils.getDateTimeForDisplay((ZonedDateTime.now())), getCommandLine());
+        } else {
+            return new HashSet<>();
         }
-        return gatkToolHeaderLines;
     }
 
     /**

diff --git a/src/main/java/org/broadinstitute/hellbender/engine/spark/GATKSparkTool.java b/src/main/java/org/broadinstitute/hellbender/engine/spark/GATKSparkTool.java
@@ -3,7 +3,6 @@
 import htsjdk.samtools.SAMFileHeader;
 import htsjdk.samtools.SAMSequenceDictionary;
 import htsjdk.variant.vcf.VCFHeaderLine;
-import htsjdk.variant.vcf.VCFSimpleHeaderLine;
 import org.apache.spark.api.java.JavaRDD;
 import org.apache.spark.api.java.JavaSparkContext;
 import org.broadinstitute.barclay.argparser.Argument;
@@ -14,12 +13,12 @@
 import org.broadinstitute.hellbender.cmdline.GATKPlugin.GATKReadFilterPluginDescriptor;
 import org.broadinstitute.hellbender.cmdline.StandardArgumentDefinitions;
 import org.broadinstitute.hellbender.cmdline.argumentcollections.*;
+import org.broadinstitute.hellbender.engine.FeatureDataSource;
+import org.broadinstitute.hellbender.engine.FeatureManager;
 import org.broadinstitute.hellbender.engine.GATKTool;
 import org.broadinstitute.hellbender.engine.TraversalParameters;
 import org.broadinstitute.hellbender.engine.datasources.ReferenceMultiSource;
 import org.broadinstitute.hellbender.engine.datasources.ReferenceWindowFunctions;
-import org.broadinstitute.hellbender.engine.FeatureDataSource;
-import org.broadinstitute.hellbender.engine.FeatureManager;
 import org.broadinstitute.hellbender.engine.filters.ReadFilter;
 import org.broadinstitute.hellbender.engine.filters.WellformedReadFilter;
 import org.broadinstitute.hellbender.engine.spark.datasources.ReadsSparkSink;
@@ -34,6 +33,7 @@
 import org.broadinstitute.hellbender.utils.io.IOUtils;
 import org.broadinstitute.hellbender.utils.read.GATKRead;
 import org.broadinstitute.hellbender.utils.read.ReadsWriteFormat;
+import org.broadinstitute.hellbender.utils.variant.GATKVariantContextUtils;
 
 import java.io.IOException;
 import java.time.ZonedDateTime;
@@ -410,7 +410,7 @@ public List<Class<? extends Annotation>> getDefaultVariantAnnotationGroups() {
         return Collections.emptyList();
     }
 
-    // TODO: 7/3/18 the two functions below are copy-pasted from GATKTool, and probably some refactoring can be done
+    // TODO: 7/9/18 the two function below (including the todo in comment) are copy-pasted from GATKTool, and probably some refactoring can be done
     /**
      * @return An abbreviated name of the toolkit for this tool. Subclasses may override to provide
      *         a custom toolkit name.
@@ -425,17 +425,13 @@ public List<Class<? extends Annotation>> getDefaultVariantAnnotationGroups() {
      * date and command line, otherwise an empty set.
      */
     protected Set<VCFHeaderLine> getDefaultToolVCFHeaderLines() {
-        final Set<VCFHeaderLine> gatkToolHeaderLines = new HashSet<>();
         if (addOutputVCFCommandLine) {
-            final Map<String, String> simpleHeaderLineMap = new HashMap<>(4);
-            simpleHeaderLineMap.put("ID", this.getClass().getSimpleName());
-            simpleHeaderLineMap.put("Version", getVersion());
-            simpleHeaderLineMap.put("Date", Utils.getDateTimeForDisplay((ZonedDateTime.now())));
-            simpleHeaderLineMap.put("CommandLine", getCommandLine());
-            gatkToolHeaderLines.add(new VCFHeaderLine("source", this.getClass().getSimpleName()));
-            gatkToolHeaderLines.add(new VCFSimpleHeaderLine(String.format("%sCommandLine", getToolkitShortName()), simpleHeaderLineMap));
+            return GATKVariantContextUtils
+                    .getDefaultVCFHeaderLines(getToolkitShortName(), this.getClass().getSimpleName(),
+                            getVersion(), Utils.getDateTimeForDisplay((ZonedDateTime.now())), getCommandLine());
+        } else {
+            return new HashSet<>();
         }
-        return gatkToolHeaderLines;
     }
 
     /**

diff --git a/src/main/java/org/broadinstitute/hellbender/utils/variant/GATKVariantContextUtils.java b/src/main/java/org/broadinstitute/hellbender/utils/variant/GATKVariantContextUtils.java
@@ -3,25 +3,26 @@
 import htsjdk.samtools.SAMSequenceDictionary;
 import htsjdk.samtools.util.IOUtil;
 import htsjdk.samtools.util.Locatable;
-import htsjdk.tribble.AbstractFeatureReader;
 import htsjdk.tribble.TribbleException;
 import htsjdk.variant.variantcontext.*;
 import htsjdk.variant.variantcontext.writer.Options;
 import htsjdk.variant.variantcontext.writer.VariantContextWriter;
 import htsjdk.variant.variantcontext.writer.VariantContextWriterBuilder;
 import htsjdk.variant.vcf.VCFConstants;
 import htsjdk.variant.vcf.VCFHeaderLine;
+import htsjdk.variant.vcf.VCFSimpleHeaderLine;
 import htsjdk.variant.vcf.VCFStandardHeaderLines;
 import org.apache.commons.io.FilenameUtils;
 import org.apache.commons.lang3.ArrayUtils;
 import org.apache.commons.lang3.tuple.MutablePair;
 import org.apache.commons.lang3.tuple.Pair;
 import org.apache.logging.log4j.LogManager;
 import org.apache.logging.log4j.Logger;
-import org.broadinstitute.hellbender.tools.walkers.genotyper.*;
+import org.broadinstitute.hellbender.tools.walkers.genotyper.GenotypeAlleleCounts;
+import org.broadinstitute.hellbender.tools.walkers.genotyper.GenotypeAssignmentMethod;
+import org.broadinstitute.hellbender.tools.walkers.genotyper.GenotypeLikelihoodCalculator;
+import org.broadinstitute.hellbender.tools.walkers.genotyper.GenotypeLikelihoodCalculators;
 import org.broadinstitute.hellbender.utils.*;
-import org.broadinstitute.hellbender.utils.collections.Permutation;
-import org.broadinstitute.hellbender.utils.genotyper.IndexedAlleleList;
 import org.broadinstitute.hellbender.utils.param.ParamUtils;
 
 import java.io.File;
@@ -49,6 +50,23 @@ public static boolean isInformative(final double[] gls) {
         return MathUtils.sum(gls) < GATKVariantContextUtils.SUM_GL_THRESH_NOCALL;
     }
 
+    /**
+     * @return A set of VCF header lines containing the tool name, version, date and command line.
+     */
+    public static Set<VCFHeaderLine> getDefaultVCFHeaderLines(final String toolkitShortName, final String toolName,
+                                                              final String versionString, final String dataTime,
+                                                              final String cmdLine) {
+        final Set<VCFHeaderLine> defaultVCFHeaderLines = new HashSet<>();
+        final Map<String, String> simpleHeaderLineMap = new HashMap<>(4);
+        simpleHeaderLineMap.put("ID", toolName);
+        simpleHeaderLineMap.put("Version", versionString);
+        simpleHeaderLineMap.put("Date", dataTime);
+        simpleHeaderLineMap.put("CommandLine", cmdLine);
+        defaultVCFHeaderLines.add(new VCFHeaderLine("source", toolName));
+        defaultVCFHeaderLines.add(new VCFSimpleHeaderLine(String.format("%sCommandLine", toolkitShortName), simpleHeaderLineMap));
+        return defaultVCFHeaderLines;
+    }
+
     /**
      * Creates a VariantContextWriter whose outputFile type is based on the extension of the output file name.
      * The default options set by VariantContextWriter are cleared before applying ALLOW_MISSING_FIELDS_IN_HEADER (if