diff options
Diffstat (limited to 'scalding/src/main/scala/sandcrawler/HBaseStatusCountJob.scala')
-rw-r--r-- | scalding/src/main/scala/sandcrawler/HBaseStatusCountJob.scala | 14 |
1 files changed, 10 insertions, 4 deletions
diff --git a/scalding/src/main/scala/sandcrawler/HBaseStatusCountJob.scala b/scalding/src/main/scala/sandcrawler/HBaseStatusCountJob.scala index b1dab0e..fd0b4e2 100644 --- a/scalding/src/main/scala/sandcrawler/HBaseStatusCountJob.scala +++ b/scalding/src/main/scala/sandcrawler/HBaseStatusCountJob.scala @@ -1,18 +1,24 @@ package sandcrawler -import com.twitter.scalding.Args +import java.util.Properties + +import cascading.property.AppProps +import cascading.tuple.Fields import com.twitter.scalding._ import com.twitter.scalding.typed.TDsl._ import org.apache.hadoop.hbase.io.ImmutableBytesWritable import org.apache.hadoop.hbase.util.Bytes import parallelai.spyglass.base.JobBase +import parallelai.spyglass.hbase.HBaseConstants.SourceMode import parallelai.spyglass.hbase.HBasePipeConversions +import parallelai.spyglass.hbase.HBaseSource class HBaseStatusCountJob(args: Args) extends JobBase(args) with HBasePipeConversions { - val source = HBaseCountJob.getHBaseSource(args("hbase-table"), - args("zookeeper-hosts"), - "grobid0:status_code") + val source = HBaseCountJob.getHBaseSource( + args("hbase-table"), + args("zookeeper-hosts"), + "grobid0:status_code") val statusPipe : TypedPipe[Long] = source .read |