Run Hive snippets against distributed Hadoop.

trahasch · Jan 28, 2012 · 4893fed · 4893fed
1 parent ae3337a
commit 4893fed
Show file tree

Hide file tree

Showing 3 changed files with 7 additions and 8 deletions.
diff --git a/ch12/src/main/hive/buckets.hive b/ch12/src/main/hive/buckets.hive
@@ -32,6 +32,8 @@ SELECT * FROM users;
 
 dfs -ls /user/hive/warehouse/bucketed_users;
 
+dfs -cat /user/hive/warehouse/bucketed_users/000000_0;
+
 ! echo; # vv sample_1_in_4;
 SELECT * FROM bucketed_users
 TABLESAMPLE(BUCKET 1 OUT OF 4 ON id);

diff --git a/ch12/src/main/hive/joins.hive b/ch12/src/main/hive/joins.hive
@@ -3,7 +3,6 @@
 ! echo; # == left_outer_join;
 ! echo; # == right_outer_join;
 ! echo; # == full_outer_join;
-! echo; # == select_things_in_sales;
 ! echo; # == semi_join;
 ! echo; # == map_join;
 
@@ -46,12 +45,6 @@ SELECT sales.*, things.*
 FROM sales FULL OUTER JOIN things ON (sales.id = things.id);
 ! echo; # ^^ full_outer_join;
 
-! echo; # vv select_things_in_sales;
-SELECT *
-FROM things
-WHERE things.id IN (SELECT id from sales);
-! echo; # ^^ select_things_in_sales;
-
 ! echo; # vv semi_join;
 SELECT *
 FROM things LEFT SEMI JOIN sales ON (sales.id = things.id);

diff --git a/snippet/bin/hiver.sh b/snippet/bin/hiver.sh
@@ -1,4 +1,8 @@
 #!/bin/bash
+#
+# Usage:
+# hiver.sh
+# hiver.sh -hiveconf fs.default.name=file:/// -hiveconf mapred.job.tracker=local
 
 set -x
 set -e
@@ -15,7 +19,7 @@ mkdir -p $actual/ch12
 
 for f in ch12/src/main/hive/*.hive; do
   out=$f.output.txt
-  hive -hiveconf fs.default.name=file:/// -hiveconf mapred.job.tracker=local -S < $f 2> /dev/null \
+  hive "$@" -S < $f 2> /dev/null \
     | grep -v '^hive> $' \
     | grep -v '^$' \
     | sed -e 's|&|\&amp;|g' \