new implementation of propagation of community to result from organization that exploits the prepared info

This commit is contained in:
Miriam Baglioni 2020-04-16 18:45:22 +02:00
parent fff1e5ec39
commit 5d772e5263
1 changed files with 76 additions and 338 deletions

View File

@ -4,6 +4,8 @@ import com.fasterxml.jackson.databind.ObjectMapper;
import com.google.gson.Gson;
import eu.dnetlib.dhp.TypedRow;
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
import eu.dnetlib.dhp.orcidtoresultfromsemrel.ResultOrcidList;
import eu.dnetlib.dhp.orcidtoresultfromsemrel.SparkOrcidToResultFromSemRelJob3;
import eu.dnetlib.dhp.schema.oaf.*;
import org.apache.commons.io.IOUtils;
import org.apache.hadoop.fs.FileSystem;
@ -13,378 +15,114 @@ import org.apache.spark.api.java.JavaRDD;
import org.apache.spark.api.java.JavaSparkContext;
import org.apache.spark.sql.Encoders;
import org.apache.spark.sql.Row;
import org.apache.spark.sql.SaveMode;
import org.apache.spark.sql.SparkSession;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import scala.Tuple2;
import java.util.*;
import java.util.stream.Collectors;
import static eu.dnetlib.dhp.PropagationConstant.*;
import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkHiveSession;
public class SparkResultToCommunityFromOrganizationJob2 {
private static final Logger log = LoggerFactory.getLogger(SparkResultToCommunityFromOrganizationJob2.class);
private static final ObjectMapper OBJECT_MAPPER = new ObjectMapper();
public class SparkResultToCommunityFromOrganization {
public static void main(String[] args) throws Exception {
String jsonConfiguration = IOUtils.toString(SparkResultToCommunityFromOrganizationJob2.class
.getResourceAsStream("/eu/dnetlib/dhp/resulttocommunityfromorganization/input_communitytoresult_parameters.json"));
final ArgumentApplicationParser parser = new ArgumentApplicationParser(
jsonConfiguration);
final ArgumentApplicationParser parser = new ArgumentApplicationParser(IOUtils.toString(SparkResultToCommunityFromOrganization.class.getResourceAsStream("/eu/dnetlib/dhp/resulttocommunityfromorganization/input_communitytoresult_parameters.json")));
parser.parseArgument(args);
Boolean isSparkSessionManaged = isSparkSessionManaged(parser);
log.info("isSparkSessionManaged: {}", isSparkSessionManaged);
String inputPath = parser.get("sourcePath");
log.info("inputPath: {}", inputPath);
final String outputPath = parser.get("outputPath");
log.info("outputPath: {}", outputPath);
final String possibleupdatespath = parser.get("preparedInfoPath");
log.info("preparedInfoPath: {}", possibleupdatespath);
final String resultClassName = parser.get("resultTableName");
log.info("resultTableName: {}", resultClassName);
final String resultType = resultClassName.substring(resultClassName.lastIndexOf(".") + 1).toLowerCase();
log.info("resultType: {}", resultType);
final Boolean saveGraph = Optional
.ofNullable(parser.get("saveGraph"))
.map(Boolean::valueOf)
.orElse(Boolean.TRUE);
log.info("saveGraph: {}", saveGraph);
Class<? extends Result> resultClazz = (Class<? extends Result>) Class.forName(resultClassName);
SparkConf conf = new SparkConf();
conf.set("hive.metastore.uris", parser.get("hive_metastore_uris"));
final SparkSession spark = SparkSession
.builder()
.appName(SparkResultToCommunityFromOrganization.class.getSimpleName())
.master(parser.get("master"))
.config(conf)
.enableHiveSupport()
.getOrCreate();
final JavaSparkContext sc = new JavaSparkContext(spark.sparkContext());
final String inputPath = parser.get("sourcePath");
final String outputPath = "/tmp/provision/propagation/communitytoresult";
final OrganizationMap organizationMap = new Gson().fromJson(parser.get("organizationtoresultcommunitymap"), OrganizationMap.class);
boolean writeUpdates = TRUE.equals(parser.get("writeUpdate"));
boolean saveGraph = TRUE.equals(parser.get("saveGraph"));
System.out.println(new Gson().toJson(organizationMap));
createOutputDirs(outputPath, FileSystem.get(spark.sparkContext().hadoopConfiguration()));
JavaRDD<Relation> relations_rdd_all = sc.textFile(inputPath + "/relation")
.map(item -> new ObjectMapper().readValue(item, Relation.class));
JavaRDD<Publication> publications_rdd_all = sc.textFile(inputPath + "/publication")
.map(item -> new ObjectMapper().readValue(item, Publication.class));
JavaRDD<Dataset> dataset_rdd_all = sc.textFile(inputPath + "/dataset")
.map(item -> new ObjectMapper().readValue(item, Dataset.class));
JavaRDD<OtherResearchProduct> orp_rdd_all = sc.textFile(inputPath + "/otheresearchproduct")
.map(item -> new ObjectMapper().readValue(item, OtherResearchProduct.class));
JavaRDD<Software> software_rdd_all = sc.textFile(inputPath + "/software")
.map(item -> new ObjectMapper().readValue(item, Software.class));
org.apache.spark.sql.Dataset<Relation> relation = spark.createDataset(relations_rdd_all.rdd(), Encoders.bean(Relation.class));
relation.createOrReplaceTempView("relation");
// String query = "SELECT source, target" +
// " FROM relation " +
// " WHERE datainfo.deletedbyinference = false " +
// " AND relClass = '" + RELATION_RESULT_ORGANIZATION_REL_CLASS + "'";
//
// org.apache.spark.sql.Dataset<Row> result_organization = spark.sql(query);
// result_organization.createOrReplaceTempView("result_organization");
//
// query = "SELECT source, target " +
// "FROM relation " +
// "WHERE datainfo.deletedbyinference = false " +
// "AND relClass = '" + RELATION_REPRESENTATIVERESULT_RESULT_CLASS + "'" ;
String query = "SELECT result_organization.source, result_organization.target, org_set " +
"FROM (SELECT source, target " +
" FROM relation " +
" WHERE datainfo.deletedbyinference = false " +
" AND relClass = '" + RELATION_RESULT_ORGANIZATION_REL_CLASS + "') result_organization " +
"LEFT JOIN (SELECT source, collect_set(target) org_set " +
" FROM relation " +
" WHERE datainfo.deletedbyinference = false " +
" AND relClass = '" + RELATION_REPRESENTATIVERESULT_RESULT_CLASS + "' " +
" GROUP BY source) organization_organization " +
"ON result_organization.target = organization_organization.source ";
org.apache.spark.sql.Dataset<Row> result_organizationset = spark.sql(query);
JavaPairRDD<String,TypedRow> result_communitySet = result_organizationset.toJavaRDD().map(r -> {
String rId = r.getString(0);
List<String> orgs = r.getList(2);
String oTarget = r.getString(1);
TypedRow tp = new TypedRow();
if (organizationMap.containsKey(oTarget)) {
tp.addAll(organizationMap.get(oTarget));
runWithSparkHiveSession(conf, isSparkSessionManaged,
spark -> {
if (isTest(parser)) {
removeOutputDir(spark, outputPath);
}
try{
for (String oId : orgs) {
if (organizationMap.containsKey(oId)) {
tp.addAll(organizationMap.get(oId));
}
}
}
catch(Exception e){
//System.out.println("organizationTargetID: " + oTarget);
}
if(tp.getAccumulator() == null ){
return null;
}
tp.setSourceId(rId);
return tp;
})
.filter(tr -> !(tr==null))
.mapToPair(toPair()).cache();
if(writeUpdates){
result_communitySet.map(c->new ObjectMapper().writeValueAsString(c)).saveAsTextFile(outputPath +"/result_communityset");
}
if(saveGraph){
updatePublicationResult(publications_rdd_all, result_communitySet)
.map(p -> new ObjectMapper().writeValueAsString(p))
.saveAsTextFile(outputPath + "/publication");
updateDatasetResult(dataset_rdd_all, result_communitySet)
.map(p -> new ObjectMapper().writeValueAsString(p))
.saveAsTextFile(outputPath + "/dataset");
updateORPResult(orp_rdd_all, result_communitySet)
.map(p -> new ObjectMapper().writeValueAsString(p))
.saveAsTextFile(outputPath + "/otherresearchproduct");
updateSoftwareResult(software_rdd_all, result_communitySet)
.map(p -> new ObjectMapper().writeValueAsString(p))
.saveAsTextFile(outputPath + "/software");
}
//relations between organziations and results
// JavaPairRDD<String, TypedRow> organization_result = relations
// .filter(r -> !r.getDataInfo().getDeletedbyinference())
// .filter(r -> RELATION_RESULT_ORGANIZATION_REL_CLASS.equals(r.getRelClass())
// && RELATION_RESULTORGANIZATION_REL_TYPE.equals(r.getRelType()))
// .map(r -> {
// TypedRow tp = new TypedRow();
// tp.setSourceId(r.getTarget());
// tp.setTargetId(r.getSource() );
// return tp;
// })
// .mapToPair(toPair());
//relations between representative organization and merged Id. One relation per merged organization
// JavaPairRDD<String, TypedRow> organization_organization = relations
// .filter(r -> !r.getDataInfo().getDeletedbyinference())
// .filter( r -> RELATION_ORGANIZATIONORGANIZATION_REL_TYPE.equals(r.getRelType())
// && RELATION_REPRESENTATIVERESULT_RESULT_CLASS.equals(r.getRelClass()))
// .map(r -> {
// TypedRow tp = new TypedRow();
// tp.setSourceId(r.getSource());
// tp.setTargetId(r.getTarget());
// return tp;
// })
// .mapToPair(toPair());
//get the original id of the organizations to be checked against the id associated to the communities
// JavaPairRDD<String, TypedRow> organizationoriginal_result = organization_result.leftOuterJoin(organization_organization)
// .map(c -> {
// if (!c._2()._2().isPresent())
// return c._2()._1();
// return c._2()._1().setSourceId(c._2()._2().get().getTargetId());
// })
// .mapToPair(toPair());
//associates to each result connected to an organization the list of communities related to that organization
// JavaPairRDD<String, TypedRow> to_add_result_communities = organizationoriginal_result.map(o -> {
// List<String> communityList = organizationMap.get(o._1());
// if (communityList.size() == 0)
// return null;
// TypedRow tp = o._2();
// tp.setAccumulator(new HashSet<>(communityList));
// tp.setSourceId(tp.getTargetId());
// return tp;
// })
// .filter(r -> !(r == null))
// .mapToPair(toPair())
// .reduceByKey((a, b) -> {
// if (a == null)
// return b;
// if (b == null)
// return a;
// a.addAll(b.getAccumulator());
// return a;
// });
// JavaRDD<Publication> publications = sc.sequenceFile(inputPath + "/publication", Text.class, Text.class)
// .map(item -> new ObjectMapper().readValue(item._2().toString(), Publication.class)).cache();
// JavaRDD<Dataset> datasets = sc.sequenceFile(inputPath + "/dataset", Text.class, Text.class)
// .map(item -> new ObjectMapper().readValue(item._2().toString(), Dataset.class)).cache();
// JavaRDD<Software> software = sc.sequenceFile(inputPath + "/software", Text.class, Text.class)
// .map(item -> new ObjectMapper().readValue(item._2().toString(), Software.class)).cache();
// JavaRDD<OtherResearchProduct> other = sc.sequenceFile(inputPath + "/otherresearchproduct", Text.class, Text.class)
// .map(item -> new ObjectMapper().readValue(item._2().toString(), OtherResearchProduct.class)).cache();
//
// JavaPairRDD<String, Result> pubs = publications.mapToPair(p -> new Tuple2<>(p.getId(),p));
// JavaPairRDD<String, Result> dss = datasets.mapToPair(p -> new Tuple2<>(p.getId(),p));
// JavaPairRDD<String, Result> sfw = software.mapToPair(p -> new Tuple2<>(p.getId(),p));
// JavaPairRDD<String, Result> orp = other.mapToPair(p -> new Tuple2<>(p.getId(),p));
//
// updateResultForCommunity(pubs, to_add_result_communities, outputPath, "publication", PROPAGATION_RESULT_COMMUNITY_ORGANIZATION_CLASS_ID, PROPAGATION_RESULT_COMMUNITY_ORGANIZATION_CLASS_NAME);
// updateResultForCommunity(dss, to_add_result_communities, outputPath, "dataset", PROPAGATION_RESULT_COMMUNITY_ORGANIZATION_CLASS_ID, PROPAGATION_RESULT_COMMUNITY_ORGANIZATION_CLASS_NAME);
// updateResultForCommunity(sfw, to_add_result_communities, outputPath, "software", PROPAGATION_RESULT_COMMUNITY_ORGANIZATION_CLASS_ID, PROPAGATION_RESULT_COMMUNITY_ORGANIZATION_CLASS_NAME);
// updateResultForCommunity(orp, to_add_result_communities, outputPath, "otherresearchproduct", PROPAGATION_RESULT_COMMUNITY_ORGANIZATION_CLASS_ID, PROPAGATION_RESULT_COMMUNITY_ORGANIZATION_CLASS_NAME);
execPropagation(spark, inputPath, outputPath, resultClazz, possibleupdatespath);
});
}
private static JavaRDD<Publication> updatePublicationResult(JavaRDD<Publication> result, JavaPairRDD<String, TypedRow> result_communitySet){
JavaRDD<Result> tmp = result.map(r -> (Result)r);
return updateResultDataset(tmp, result_communitySet).map(r -> (Publication)r);
private static <R extends Result> void execPropagation(SparkSession spark, String inputPath, String outputPath,
Class<R> resultClazz, String possibleUpdatesPath) {
org.apache.spark.sql.Dataset<ResultCommunityList> possibleUpdates = readResultCommunityList(spark, possibleUpdatesPath);
org.apache.spark.sql.Dataset<R> result = readPathEntity(spark, inputPath, resultClazz);
}
private static JavaRDD<Dataset> updateDatasetResult(JavaRDD<Dataset> result, JavaPairRDD<String, TypedRow> result_communitySet){
JavaRDD<Result> tmp = result.map(r -> (Result)r);
return updateResultDataset(tmp, result_communitySet).map(r -> (Dataset) r);
}
private static JavaRDD<OtherResearchProduct> updateORPResult(JavaRDD<OtherResearchProduct> result, JavaPairRDD<String, TypedRow> result_communitySet){
JavaRDD<Result> tmp = result.map(r -> (Result)r);
return updateResultDataset(tmp, result_communitySet).map(r -> (OtherResearchProduct)r);
}
private static JavaRDD<Software> updateSoftwareResult(JavaRDD<Software> result, JavaPairRDD<String, TypedRow> result_communitySet){
JavaRDD<Result> tmp = result.map(r -> (Result)r);
return updateResultDataset(tmp, result_communitySet).map(r -> (Software) r);
}
private static JavaRDD<Result> updateResultDataset(JavaRDD<Result> result, JavaPairRDD<String, TypedRow> result_communitySet){
return result.mapToPair(p -> new Tuple2<>(p.getId(),p)).leftOuterJoin(result_communitySet)
.map(c -> {
Result r = c._2()._1();
if(c._2()._2().isPresent()){
Set<String> communitySet = c._2()._2().get().getAccumulator();
List<String> contextList = r.getContext().stream().map(con -> con.getId()).collect(Collectors.toList());
result
.joinWith(possibleUpdates, result.col("id").equalTo(possibleUpdates.col("resultId")),
"left_outer")
.map(value -> {
R ret = value._1();
Optional<ResultCommunityList> rcl = Optional.ofNullable(value._2());
if(rcl.isPresent()){
ArrayList<String> communitySet = rcl.get().getCommunityList();
List<String> contextList = ret.getContext().stream().map(con -> con.getId()).collect(Collectors.toList());
for(String cId:communitySet){
if(!contextList.contains(cId)){
Context newContext = new Context();
newContext.setId(cId);
newContext.setDataInfo(Arrays.asList(getDataInfo(PROPAGATION_DATA_INFO_TYPE,
PROPAGATION_RESULT_COMMUNITY_ORGANIZATION_CLASS_ID, PROPAGATION_RESULT_COMMUNITY_ORGANIZATION_CLASS_NAME)));
r.getContext().add(newContext);
ret.getContext().add(newContext);
}
}
}
return r;
});
}
// private static void updateResult(JavaPairRDD<String, Result> results, JavaPairRDD<String, TypedRow> toupdateresult, String outputPath, String type) {
// results.leftOuterJoin(toupdateresult)
// .map(p -> {
// Result r = p._2()._1();
// if (p._2()._2().isPresent()){
// Set<String> communityList = p._2()._2().get().getAccumulator();
// for(Context c: r.getContext()){
// if (communityList.contains(c.getId())){
// //verify if the datainfo for this context contains propagation
// if (!c.getDataInfo().stream().map(di -> di.getInferenceprovenance()).collect(Collectors.toSet()).contains(PROPAGATION_DATA_INFO_TYPE)){
// c.getDataInfo().add(getDataInfo(PROPAGATION_DATA_INFO_TYPE, PROPAGATION_RESULT_COMMUNITY_ORGANIZATION_CLASS_ID, PROPAGATION_RESULT_COMMUNITY_ORGANIZATION_CLASS_NAME));
// //community id already in the context of the result. Remove it from the set that has to be added
// communityList.remove(c.getId());
// }
// }
// }
// List<Context> cc = r.getContext();
// for(String cId: communityList){
// Context context = new Context();
// context.setId(cId);
// context.setDataInfo(Arrays.asList(getDataInfo(PROPAGATION_DATA_INFO_TYPE, PROPAGATION_RESULT_COMMUNITY_ORGANIZATION_CLASS_ID, PROPAGATION_RESULT_COMMUNITY_ORGANIZATION_CLASS_NAME)));
// cc.add(context);
// }
// r.setContext(cc);
// }
// return r;
// })
// .map(p -> new ObjectMapper().writeValueAsString(p))
// .saveAsTextFile(outputPath+"/"+type);
// }
return ret;
}, Encoders.bean(resultClazz))
.toJSON()
.write()
.mode(SaveMode.Overwrite)
.option("compression","gzip")
.text(outputPath);
}
/*
package eu.dnetlib.data.mapreduce.hbase.propagation.communitythroughorganization;
import com.google.gson.Gson;
import eu.dnetlib.data.mapreduce.hbase.propagation.Value;
import eu.dnetlib.data.mapreduce.util.OafRowKeyDecoder;
import eu.dnetlib.data.proto.OafProtos;
import eu.dnetlib.data.proto.TypeProtos;
import org.apache.commons.lang3.StringUtils;
import org.apache.hadoop.hbase.client.Result;
import org.apache.hadoop.hbase.io.ImmutableBytesWritable;
import org.apache.hadoop.hbase.mapreduce.TableMapper;
import org.apache.hadoop.hbase.util.Bytes;
import org.apache.hadoop.io.Text;
import java.io.IOException;
import java.util.Set;
import static eu.dnetlib.data.mapreduce.hbase.propagation.PropagationConstants.*;
import static eu.dnetlib.data.mapreduce.hbase.propagation.Utils.getEntity;
import static eu.dnetlib.data.mapreduce.hbase.propagation.Utils.getRelationTarget;
public class PropagationCommunityThroughOrganizationMapper extends TableMapper<ImmutableBytesWritable, Text> {
private Text valueOut;
private ImmutableBytesWritable keyOut;
private OrganizationMap organizationMap;
//seleziono il tipo della entry:
//Result:
//se non e' deleted by inference ed ha organizzazioni a cui e' associato,
// //emetto id della relazione ed id del risultato per ogni organizzazione a cui e' associato
//ORGANIZATION:
//se non e' deleted by inference e non e' deduplicata emetto l'id della organizzazione
//se non e' deleted by inference ed e' deduplicata, emetto id della organizzazione ed id del deduplicato e lista delle organizzazioni mergiate
@Override
protected void setup(final Context context) throws IOException, InterruptedException {
super.setup(context);
valueOut = new Text();
keyOut = new ImmutableBytesWritable();
organizationMap = new Gson().fromJson(context.getConfiguration().get("organizationtoresult.community.map"), OrganizationMap.class);
System.out.println("got organizationtoresult map: " + new Gson().toJson(organizationMap)) ;
}
@Override
protected void map(final ImmutableBytesWritable keyIn, final Result value, final Context context) throws IOException, InterruptedException {
final TypeProtos.Type type = OafRowKeyDecoder.decode(keyIn.copyBytes()).getType();
final OafProtos.OafEntity entity = getEntity(value, type);//getEntity already verified that it is not delByInference
if (entity != null) {
switch (type) {
case organization:
DedupedList communityList = getCommunityList(Bytes.toString(keyIn.get()),
getRelationTarget(value, DEDUP_RELATION_ORGANIZATION + REL_DEDUP_REPRESENTATIVE_RESULT, context, COUNTER_PROPAGATION), context);
if (communityList.size() > 0){
valueOut.set(Value.newInstance(
new Gson().toJson(
communityList, //search for organizationtoresult it merges
DedupedList.class),
ORGANIZATION_COMMUNITY_TRUST,
Type.fromorganization).toJson());
context.write(keyIn, valueOut);
context.getCounter(COUNTER_PROPAGATION, "emit for organizationtoresult ").increment(1);
}else{
context.getCounter(COUNTER_PROPAGATION, "community list size = 0 ").increment(1);
}
break;
case result:
Set<String> result_organization = getRelationTarget(value, RELATION_ORGANIZATION + REL_RESULT_ORGANIZATION, context, COUNTER_PROPAGATION);
for(String org: result_organization)
emit(org, Bytes.toString(keyIn.get()), context);
break;
}
}
}
private DedupedList getCommunityList(String organizationId, Set<String> relationTarget, Context context) {
DedupedList communityList = new DedupedList();
relationTarget.stream().forEach(org -> communityList.addAll(organizationMap.get(StringUtils.substringAfter(org, "|"))));
communityList.addAll(organizationMap.get(StringUtils.substringAfter(organizationId,"|")));
communityList.stream().forEach(c->context.getCounter(COUNTER_PROPAGATION,"found organizationtoresult for " + c).increment(1));
return communityList;
}
private void emit(String org, String resId, Context context) throws IOException, InterruptedException {
keyOut.set(Bytes.toBytes(org));
valueOut.set(Value.newInstance(resId, ORGANIZATION_COMMUNITY_TRUST, Type.fromresult).toJson());
context.write(keyOut,valueOut);
context.getCounter(COUNTER_PROPAGATION, "emit for result").increment(1);
private static org.apache.spark.sql.Dataset<ResultCommunityList> readResultCommunityList(SparkSession spark, String possibleUpdatesPath) {
return spark
.read()
.textFile(possibleUpdatesPath)
.map(value -> OBJECT_MAPPER.readValue(value, ResultCommunityList.class), Encoders.bean(ResultCommunityList.class));
}
}
*/