diff options
Diffstat (limited to 'pig/hbase-count-rows.pig')
| -rw-r--r-- | pig/hbase-count-rows.pig | 15 |
1 files changed, 15 insertions, 0 deletions
diff --git a/pig/hbase-count-rows.pig b/pig/hbase-count-rows.pig new file mode 100644 index 0000000..57e83a5 --- /dev/null +++ b/pig/hbase-count-rows.pig @@ -0,0 +1,15 @@ + +REGISTER /usr/lib/hbase/lib/hbase-client-0.98.6-cdh5.3.1.jar +REGISTER /usr/lib/hbase/lib/hbase-common-0.98.6-cdh5.3.1.jar +REGISTER /usr/lib/hbase/lib/hbase-hadoop2-compat-0.98.6-cdh5.3.1.jar +REGISTER /usr/lib/hbase/lib/hbase-protocol-0.98.6-cdh5.3.1.jar + +set hbase.zookeeper.quorum 'mtrcs-zk1.us.archive.org,mtrcs-zk2.us.archive.org,mtrcs-zk3.us.archive.org' + +data = LOAD 'hbase://wbgrp-journal-extract-0-qa' + USING org.apache.pig.backend.hadoop.hbase.HBaseStorage('grobid0:status_code', '-loadKey true') + AS (key:CHARARRAY, status:CHARARRAY); + +data_group = GROUP data ALL; +data_count = FOREACH data_group GENERATE COUNT(data); +DUMP data_count; |