forked from D-Net/dnet-hadoop
removed the repartition(1). Added code for the creation of an archive containing all the parts dumped for each community
This commit is contained in:
parent
1991a49f70
commit
341dc59ead
|
@ -59,7 +59,6 @@ public class CommunitySplit implements Serializable {
|
||||||
try {
|
try {
|
||||||
community_products.first();
|
community_products.first();
|
||||||
community_products
|
community_products
|
||||||
.repartition(1)
|
|
||||||
.write()
|
.write()
|
||||||
.option("compression", "gzip")
|
.option("compression", "gzip")
|
||||||
.mode(SaveMode.Overwrite)
|
.mode(SaveMode.Overwrite)
|
||||||
|
|
Loading…
Reference in New Issue