Skip to content

Commit

Permalink
[ADAM-1679] Unmapped items should not get caught in requirement when …
Browse files Browse the repository at this point in the history
…sorting.

Resolves #1679.
  • Loading branch information
fnothaft committed Aug 27, 2017
1 parent 9b51df5 commit f365b3f
Show file tree
Hide file tree
Showing 2 changed files with 21 additions and 2 deletions.
Expand Up @@ -340,7 +340,7 @@ trait GenomicRDD[T, U <: GenomicRDD[T, U]] extends Logging {

// We don't use ValidationStringency here because multimapped elements
// break downstream methods.
require(coveredRegions.size == 1,
require(coveredRegions.size <= 1,
"Cannot sort RDD containing a multimapped element. %s covers %s.".format(
elem, coveredRegions.mkString(",")))

Expand Down Expand Up @@ -403,7 +403,7 @@ trait GenomicRDD[T, U <: GenomicRDD[T, U]] extends Logging {

// We don't use ValidationStringency here because multimapped elements
// break downstream methods.
require(coveredRegions.size == 1,
require(coveredRegions.size <= 1,
"Cannot sort RDD containing a multimapped element. %s covers %s.".format(
elem, coveredRegions.mkString(",")))

Expand Down
Expand Up @@ -1380,4 +1380,23 @@ class AlignmentRecordRDDSuite extends ADAMFunSuite {
assert(htsjdkPg.getProgramVersion === "1")
assert(htsjdkPg.getPreviousProgramGroupId === "ppg")
}

sparkTest("GenomicRDD.sort does not fail on unmapped reads") {
val inputPath = testFile("unmapped.sam")
val reads: AlignmentRecordRDD = sc.loadAlignments(inputPath)
assert(reads.rdd.count === 200)

val sorted = reads.sort(stringency = ValidationStringency.SILENT)
assert(sorted.rdd.count === 102)
}

sparkTest("GenomicRDD.sortLexicographically does not fail on unmapped reads") {
val inputPath = testFile("unmapped.sam")
val reads: AlignmentRecordRDD = sc.loadAlignments(inputPath)
assert(reads.rdd.count === 200)

val sorted = reads.sortLexicographically(
stringency = ValidationStringency.SILENT)
assert(sorted.rdd.count === 102)
}
}

0 comments on commit f365b3f

Please sign in to comment.