wikipedia

drop table wikipedia;
create external table wikipedia(
  domain string,
  page string,
  count bigint,
  size bigint
)
partitioned by (datewiki int)
row format delimited fields terminated by ' ' 
stored as textfile 
;

vi wikiday.sh
echo "start $1 $2 $3"
echo "creating hdfs dir..."
sudo -u hive hadoop fs -mkdir /test/wikipedia/datewiki=$1$2$3
echo "done creating hdfs "
for i in $(seq -f "%02g" 00 23)
do
  echo "start download ${i}..."
  sudo -u hive sh -c "wget -O - -o /dev/null https://dumps.wikimedia.org/other/pagecounts-raw/$1/$1-$2/pagecounts-$1$2$3-${i}0000.gz | hadoop fs -put - /test/wikipedia/datewiki=$1$2$3/${i}0000.gz"
  echo "done download ${i}"
done
echo "adding hive partition"
beeline -u "jdbc:hive2://localhost:10000/default" -n "admin" -e "alter table wikipedia add partition (datewiki=$1$2$3) location '/test/wikipedia/datewiki=$1$2$3/';"
echo "done $1 $2 $3"

vi wiki.sh
d=2015-01-01
while [ "$d" != 2015-01-04 ]; do 
  echo "${d:0:4}" "${d:5:2}" "${d:8:2}"
  ./wikiday.sh "${d:0:4}" "${d:5:2}" "${d:8:2}"
  d=$(date -I -d "$d + 1 day")
done