make col counter generic

2 files changed, 35 insertions, 6 deletions

diff --git a/please b/please
index 0cc4b72..8a46978 100755
--- a/please
+++ b/please
@@ -165,6 +165,30 @@ def run_grobidscorabledump(args):
             env=args.env)
     subprocess.call(cmd, shell=True)
 
+def run_colcount(args):
+    if args.rebuild:
+        rebuild_scalding()
+    print("Starting colcount job...")
+    output = "{}/output-{}/{}-colcount-{}".format(
+        HDFS_DIR,
+        args.env,
+        datetime.strftime(datetime.now(), "%Y-%m-%d-%H%M.%S"),
+        args.column)
+    cmd = """hadoop jar \
+        scalding/target/scala-2.11/sandcrawler-assembly-0.2.0-SNAPSHOT.jar \
+        com.twitter.scalding.Tool sandcrawler.HBaseColCountJob \
+        --hdfs \
+        --app.conf.path scalding/ia_cluster.conf \
+        --hbase-table wbgrp-journal-extract-0-{env} \
+        --zookeeper-hosts {zookeeper_hosts} \
+        --column {column} \
+        --output {output}""".format(
+            column=args.column,
+            output=output,
+            zookeeper_hosts=ZOOKEEPER_HOSTS,
+            env=args.env)
+    subprocess.call(cmd, shell=True)
+
 def main():
     parser = argparse.ArgumentParser()
 
@@ -206,6 +230,10 @@ def main():
     sub_grobidscorabledump = subparsers.add_parser('grobid-scorable-dump')
     sub_grobidscorabledump.set_defaults(func=run_grobidscorabledump)
 
+    sub_colcount = subparsers.add_parser('col-count')
+    sub_colcount.set_defaults(func=run_colcount)
+    sub_colcount.add_argument('column',
+        help="column name to use in count")
 
     args = parser.parse_args()
     if not args.__dict__.get("func"):
diff --git a/scalding/src/main/scala/sandcrawler/GrobidMetadataCountJob.scala b/scalding/src/main/scala/sandcrawler/HBaseColCountJob.scala
index 08f3340..a007339 100644
--- a/scalding/src/main/scala/sandcrawler/GrobidMetadataCountJob.scala
+++ b/scalding/src/main/scala/sandcrawler/HBaseColCountJob.scala
@@ -10,27 +10,28 @@ import parallelai.spyglass.hbase.HBaseConstants.SourceMode
 import parallelai.spyglass.hbase.HBasePipeConversions
 import parallelai.spyglass.hbase.HBaseSource
 
-class GrobidMetadataCountJob(args: Args) extends JobBase(args) with HBasePipeConversions {
+class HBaseColCountJob(args: Args) extends JobBase(args) with HBasePipeConversions {
 
   val output = args("output")
 
-  GrobidMetadataCountJob.getHBaseSource(
+  HBaseColCountJob.getHBaseSource(
     args("hbase-table"),
-    args("zookeeper-hosts"))
+    args("zookeeper-hosts"),
+    args("column")
     .read
     .debug
     .groupAll { _.size('count) }
     .write(Tsv(output))
 }
 
-object GrobidMetadataCountJob {
+object HBaseColCountJob {
 
   // eg, "wbgrp-journal-extract-0-qa",7 "mtrcs-zk1.us.archive.org:2181"
-  def getHBaseSource(hbaseTable: String, zookeeperHosts: String) : HBaseSource = {
+  def getHBaseSource(hbaseTable: String, zookeeperHosts: String, col: String) : HBaseSource = {
     HBaseBuilder.build(
       hbaseTable,
       zookeeperHosts,
-      List("grobid0:metadata"),
+      List(col),
       SourceMode.SCAN_ALL)
   }
 }