aboutsummaryrefslogtreecommitdiffstats
path: root/scalding/src/main/scala/sandcrawler/MatchBenchmarkJob.scala
diff options
context:
space:
mode:
authorBryan Newbold <bnewbold@archive.org>2018-08-24 12:19:09 -0700
committerBryan Newbold <bnewbold@archive.org>2018-08-24 12:19:09 -0700
commitf50d4e081f7994a167c4974ee9d3f6e1f8eae478 (patch)
tree00cf69ffe345f7766e7477cb9b1f5f7448b4e4fb /scalding/src/main/scala/sandcrawler/MatchBenchmarkJob.scala
parent344531eb6a5cdd4ea15e4d82050368c5af0eafee (diff)
parent5340caad7b39ad29bba77d2a3e486db7a6b1977b (diff)
downloadsandcrawler-f50d4e081f7994a167c4974ee9d3f6e1f8eae478.tar.gz
sandcrawler-f50d4e081f7994a167c4974ee9d3f6e1f8eae478.zip
Merge branch 'bnewbold-match-quality'
Manually resolved merge conflict in: please
Diffstat (limited to 'scalding/src/main/scala/sandcrawler/MatchBenchmarkJob.scala')
-rw-r--r--scalding/src/main/scala/sandcrawler/MatchBenchmarkJob.scala29
1 files changed, 29 insertions, 0 deletions
diff --git a/scalding/src/main/scala/sandcrawler/MatchBenchmarkJob.scala b/scalding/src/main/scala/sandcrawler/MatchBenchmarkJob.scala
new file mode 100644
index 0000000..1578258
--- /dev/null
+++ b/scalding/src/main/scala/sandcrawler/MatchBenchmarkJob.scala
@@ -0,0 +1,29 @@
+package sandcrawler
+
+import cascading.pipe.Pipe
+import com.twitter.scalding.Args
+import com.twitter.scalding.TypedPipe
+import com.twitter.scalding.TypedTsv
+import parallelai.spyglass.base.JobBase
+
+class MatchBenchmarkJob(args: Args) extends JobBase(args) {
+ // TODO: Instantiate any subclass of Scorable specified in args.
+ val sc1 : Scorable = new BibjsonScorable()
+ val sc2 : Scorable = new BibjsonScorable()
+ val leftArgs = args + ("bibjson-input" -> List(args("left-bibjson")))
+ val rightArgs = args + ("bibjson-input" -> List(args("right-bibjson")))
+ val pipe1 : TypedPipe[(String, ReduceFeatures)] = sc1.getInputPipe(leftArgs)
+ val pipe2 : TypedPipe[(String, ReduceFeatures)] = sc2.getInputPipe(rightArgs)
+
+ pipe1.join(pipe2).map { entry =>
+ val (slug : String, (features1 : ReduceFeatures, features2 : ReduceFeatures)) = entry
+ new ReduceOutput(
+ slug,
+ Scorable.computeSimilarity(features1, features2),
+ features1.json,
+ features2.json)
+ }
+ //TypedTsv doesn't work over case classes.
+ .map { entry => (entry.slug, entry.score, entry.json1, entry.json2) }
+ .write(TypedTsv[(String, Int, String, String)](args("output")))
+}