aboutsummaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorEllen Spertus <ellen.spertus@gmail.com>2018-06-06 09:40:04 -0700
committerEllen Spertus <ellen.spertus@gmail.com>2018-06-06 09:40:04 -0700
commitcc54dea9c77f8386bafd71977b2be1b3b21f0aae (patch)
tree6f61a189276ea99c4a15dde55969155ba619809f
parente746171526f578789c55d7aa690cc8381e139439 (diff)
downloadsandcrawler-cc54dea9c77f8386bafd71977b2be1b3b21f0aae.tar.gz
sandcrawler-cc54dea9c77f8386bafd71977b2be1b3b21f0aae.zip
Added job and test for counting mime types.
-rw-r--r--scalding/src/main/scala/sandcrawler/HBaseMimeCountJob.scala28
-rw-r--r--scalding/src/test/scala/sandcrawler/HBaseMimeCountTest.scala68
2 files changed, 96 insertions, 0 deletions
diff --git a/scalding/src/main/scala/sandcrawler/HBaseMimeCountJob.scala b/scalding/src/main/scala/sandcrawler/HBaseMimeCountJob.scala
new file mode 100644
index 0000000..819a652
--- /dev/null
+++ b/scalding/src/main/scala/sandcrawler/HBaseMimeCountJob.scala
@@ -0,0 +1,28 @@
+package sandcrawler
+
+import cascading.property.AppProps
+import cascading.tuple.Fields
+import com.twitter.scalding._
+import java.util.Properties
+import parallelai.spyglass.base.JobBase
+import parallelai.spyglass.hbase.{HBaseSource, HBasePipeConversions}
+import parallelai.spyglass.hbase.HBaseConstants.SourceMode
+
+class HBaseMimeCountJob(args: Args) extends JobBase(args) with HBasePipeConversions {
+ val output = args("output")
+
+ HBaseMimeCountJob.getHBaseSource
+ .read
+ .fromBytesWritable(List('mime))
+ .debug
+ .groupBy('mime){group => group.size('count)}
+ .write(Tsv(output))
+}
+
+object HBaseMimeCountJob {
+ def getHBaseSource = HBaseBuilder.build(
+ "wbgrp-journal-extract-0-qa", // HBase Table Name
+ "mtrcs-zk1.us.archive.org:2181", // HBase Zookeeper server (to get runtime config info; can be array?)
+ List("file:mime"),
+ SourceMode.SCAN_ALL)
+}
diff --git a/scalding/src/test/scala/sandcrawler/HBaseMimeCountTest.scala b/scalding/src/test/scala/sandcrawler/HBaseMimeCountTest.scala
new file mode 100644
index 0000000..000420b
--- /dev/null
+++ b/scalding/src/test/scala/sandcrawler/HBaseMimeCountTest.scala
@@ -0,0 +1,68 @@
+package sandcrawler
+
+import cascading.tuple.{Tuple, Fields}
+import com.twitter.scalding.{JobTest, Tsv, TupleConversions}
+import org.apache.hadoop.hbase.io.ImmutableBytesWritable
+import org.apache.hadoop.hbase.util.Bytes
+import org.junit.runner.RunWith
+import org.scalatest.FunSpec
+import org.scalatest.junit.JUnitRunner
+import org.slf4j.LoggerFactory
+import parallelai.spyglass.hbase.HBaseSource
+import parallelai.spyglass.hbase.HBaseConstants.SourceMode
+import scala._
+
+/**
+ * Example of how to define tests for HBaseSource
+ */
+@RunWith(classOf[JUnitRunner])
+class HBaseMimeCountTest extends FunSpec with TupleConversions {
+
+ val output = "/tmp/testOutput"
+
+ val log = LoggerFactory.getLogger(this.getClass.getName)
+
+ val mimeType1 = "text/html"
+ val mimeType2 = "application/pdf"
+
+ val sampleData = List(
+ List("sha1:K2DKSSVTXWPRMFDTWSTCQW3RVWRIOV3Q", mimeType1),
+ List("sha1:C3YNNEGH5WAG5ZAAXWAEBNXJWT6CZ3WU", mimeType1),
+ List("sha1:SDKUVHC3YNNEGH5WAG5ZAAXWAEBNX4WT", mimeType2),
+ List("sha1:35985C3YNNEGH5WAG5ZAAXWAEBNXJW56", mimeType2),
+ List("sha1:885C3YNNEGH5WAG5ZAAXWA8BNXJWT6CZ", mimeType2),
+ List("sha1:00904C3YNNEGH5WAG5ZA9XWAEBNXJWT6", mimeType2),
+ List("sha1:249C3YNNEGH5WAG5ZAAXWAEBNXJWT6CZ", mimeType1),
+ List("sha1:095893C3YNNEGH5WAG5ZAAXWAEBNXJWT", mimeType2)
+ )
+
+ val mimeType1Count = 3
+ val mimeType2Count = 5
+
+ JobTest("sandcrawler.HBaseMimeCountJob")
+ .arg("test", "")
+ .arg("app.conf.path", "app.conf")
+ .arg("output", output)
+ .arg("debug", "true")
+ .source[Tuple](HBaseMimeCountJob.getHBaseSource,
+ sampleData.map(l => new Tuple(l.map(s => {new ImmutableBytesWritable(Bytes.toBytes(s))}):_*)))
+ .sink[Tuple](Tsv(output)) {
+ outputBuffer =>
+ it("should return a 2-element list.") {
+ println("outputBuffer.size => " + outputBuffer.size)
+ println("outputBuffer(0) => " + outputBuffer(0))
+ println("outputBuffer(1) => " + outputBuffer(1))
+ assert(outputBuffer.size === 2)
+ }
+
+ val counts = outputBuffer.map(t => (t.getString(0), t.getInteger(1))).toMap
+
+ it("should have the appropriate number of each mime type") {
+ assert(counts(mimeType1) == mimeType1Count)
+ assert(counts(mimeType2) == mimeType2Count)
+ }
+ }
+ .run
+ .finish
+
+}