lsmyrnaios
/
dnet-hadoop
forked from D-Net/dnet-hadoop
1
0
Fork 0
Code Pull Requests Activity

extended ebiLinks to create a dataset before generation of OAF

This commit is contained in:
Sandro La Bruzzo 2021-07-08 10:26:21 +02:00
parent a4a54a3786
commit 4d53402712
1 changed files with 7 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
dhp-workflows/dhp-graph-mapper/src/main/java/eu/dnetlib/dhp/sx/graph/ebi/SparkEBILinksToOaf.scala
View File

@ -2,10 +2,12 @@ package eu.dnetlib.dhp.sx.graph.ebi
import eu.dnetlib.dhp.application.ArgumentApplicationParser import eu.dnetlib.dhp.application.ArgumentApplicationParser
import eu.dnetlib.dhp.schema.oaf.Oaf import eu.dnetlib.dhp.schema.oaf.Oaf
import eu.dnetlib.dhp.sx.graph.bio
import eu.dnetlib.dhp.sx.graph.bio.BioDBToOAF import eu.dnetlib.dhp.sx.graph.bio.BioDBToOAF
import eu.dnetlib.dhp.sx.graph.bio.BioDBToOAF.EBILinkItem import eu.dnetlib.dhp.sx.graph.bio.BioDBToOAF.EBILinkItem
import org.apache.commons.io.IOUtils import org.apache.commons.io.IOUtils
import org.apache.spark.SparkConf import org.apache.spark.SparkConf
import org.apache.spark.rdd.RDD
import org.apache.spark.sql.{Dataset, Encoder, Encoders, SaveMode, SparkSession} import org.apache.spark.sql.{Dataset, Encoder, Encoders, SaveMode, SparkSession}
import org.slf4j.{Logger, LoggerFactory} import org.slf4j.{Logger, LoggerFactory}
object SparkEBILinksToOaf { object SparkEBILinksToOaf {
@ -29,6 +31,11 @@ object SparkEBILinksToOaf {
import spark.implicits._ import spark.implicits._
implicit val PMEncoder: Encoder[Oaf] = Encoders.kryo(classOf[Oaf]) implicit val PMEncoder: Encoder[Oaf] = Encoders.kryo(classOf[Oaf])
val ebi_rdd:Dataset[EBILinkItem] = spark.createDataset(spark.sparkContext.textFile(sourcePath).map(s => BioDBToOAF.extractEBILinksFromDump(s))).as[EBILinkItem]
ebi_rdd.write.mode(SaveMode.Overwrite).save(s"${sourcePath}_dataset")
val ebLinks:Dataset[EBILinkItem] = spark.read.load(sourcePath).as[EBILinkItem].filter(l => l.links!= null) val ebLinks:Dataset[EBILinkItem] = spark.read.load(sourcePath).as[EBILinkItem].filter(l => l.links!= null)
ebLinks.flatMap(j =>BioDBToOAF.parse_ebi_links(j.links)) ebLinks.flatMap(j =>BioDBToOAF.parse_ebi_links(j.links))