aboutsummaryrefslogtreecommitdiffstats
path: root/scalding
diff options
context:
space:
mode:
authorBryan Newbold <bnewbold@archive.org>2018-08-19 19:20:05 -0700
committerBryan Newbold <bnewbold@archive.org>2018-08-21 20:31:45 -0700
commit6fbe71c0d36012d9096c1f5557aa64a53f6218d5 (patch)
treeb3118ca8f13d899c859efcd3fdf00427a7f5ea42 /scalding
parente99700cb521d82a750497d58bcb04d8cf1abcd80 (diff)
downloadsandcrawler-6fbe71c0d36012d9096c1f5557aa64a53f6218d5.tar.gz
sandcrawler-6fbe71c0d36012d9096c1f5557aa64a53f6218d5.zip
local bibjson-to-bibjson matching job (no tests)
Diffstat (limited to 'scalding')
-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")))
+}