building (but nullpointer) spyglass integration

1 files changed, 23 insertions, 0 deletions

diff --git a/scald-mvp/src/main/scala/sandcrawler/HBaseRowCountJob.scala b/scald-mvp/src/main/scala/sandcrawler/HBaseRowCountJob.scala
new file mode 100644
index 0000000..734abaa
--- /dev/null
+++ b/scald-mvp/src/main/scala/sandcrawler/HBaseRowCountJob.scala
@@ -0,0 +1,23 @@
+package sandcrawler
+
+import com.twitter.scalding._
+import parallelai.spyglass.hbase.{HBaseSource, HBasePipeConversions, HBaseConstants}
+
+class HBaseRowCountJob(args: Args) extends Job(args) {
+
+  // For now doesn't actually count, just dumps a "word count"
+
+  val hbs = new HBaseSource(
+    "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?)
+    'key,                             // ... then a list of column names
+    sourceMode = HBaseConstants.SourceMode.SCAN_ALL)
+/*
+    .read
+    .map { word => (word, 1L) }
+    .sumByKey
+    .write(TypedTsv[(String, Long)](args("output")))
+    // The compiler will enforce the type coming out of the sumByKey is the same as the type we have for our sink
+    .flatMap { line => line.split("\\s+") }
+*/
+}