authors and works are now updated in two separate spark actions of the wf

This commit is contained in:
Enrico Ottonello 2020-12-12 02:06:21 +01:00
parent 858efbfad1
commit efe4c2a9c5
10 changed files with 504 additions and 75 deletions

View File

@ -1,7 +1,9 @@
package eu.dnetlib.dhp.schema.orcid;
public class AuthorSummary extends OrcidData {
import java.io.Serializable;
public class AuthorSummary extends OrcidData implements Serializable {
AuthorData authorData;
AuthorHistory authorHistory;

View File

@ -1,11 +1,13 @@
package eu.dnetlib.dhp.schema.orcid;
import java.io.Serializable;
/**
* This class models the data related to external id, that are retrieved from an orcid publication
*/
public class ExternalId {
public class ExternalId implements Serializable {
private String type;
private String value;
private String relationShip;

View File

@ -1,7 +1,9 @@
package eu.dnetlib.dhp.schema.orcid;
public class OrcidData {
import java.io.Serializable;
public class OrcidData implements Serializable {
protected String base64CompressData;
protected String statusCode;
protected String downloadDate;

View File

@ -1,11 +1,13 @@
package eu.dnetlib.dhp.schema.orcid;
import java.io.Serializable;
/**
* This class models the data related to a publication date, that are retrieved from an orcid publication
*/
public class PublicationDate {
public class PublicationDate implements Serializable {
private String year;
private String month;
private String day;

View File

@ -1,7 +1,9 @@
package eu.dnetlib.dhp.schema.orcid;
public class Work extends OrcidData {
import java.io.Serializable;
public class Work extends OrcidData implements Serializable {
WorkDetail workDetail;
public WorkDetail getWorkDetail() {

View File

@ -0,0 +1,178 @@
package eu.dnetlib.doiboost.orcid;
import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkSession;
import java.io.IOException;
import java.util.Objects;
import java.util.Optional;
import org.apache.commons.io.IOUtils;
import org.apache.commons.lang3.StringUtils;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.io.compress.GzipCodec;
import org.apache.spark.SparkConf;
import org.apache.spark.api.java.JavaSparkContext;
import org.apache.spark.api.java.function.Function;
import org.apache.spark.sql.Dataset;
import org.apache.spark.sql.Encoders;
import org.apache.spark.util.LongAccumulator;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import com.fasterxml.jackson.annotation.JsonInclude;
import com.fasterxml.jackson.databind.ObjectMapper;
import com.google.gson.JsonElement;
import com.google.gson.JsonParser;
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
import eu.dnetlib.dhp.schema.orcid.AuthorSummary;
import eu.dnetlib.doiboost.orcid.xml.XMLRecordParser;
import scala.Tuple2;
public class SparkUpdateOrcidAuthors {
private static final ObjectMapper OBJECT_MAPPER = new ObjectMapper()
.setSerializationInclusion(JsonInclude.Include.NON_NULL);
public static void main(String[] args) throws IOException, Exception {
Logger logger = LoggerFactory.getLogger(SparkUpdateOrcidDatasets.class);
final ArgumentApplicationParser parser = new ArgumentApplicationParser(
IOUtils
.toString(
SparkUpdateOrcidDatasets.class
.getResourceAsStream(
"/eu/dnetlib/dhp/doiboost/download_orcid_data.json")));
parser.parseArgument(args);
Boolean isSparkSessionManaged = Optional
.ofNullable(parser.get("isSparkSessionManaged"))
.map(Boolean::valueOf)
.orElse(Boolean.TRUE);
final String workingPath = parser.get("workingPath");
// final String outputPath = parser.get("outputPath");
SparkConf conf = new SparkConf();
runWithSparkSession(
conf,
isSparkSessionManaged,
spark -> {
JavaSparkContext sc = JavaSparkContext.fromSparkContext(spark.sparkContext());
LongAccumulator oldAuthorsFoundAcc = spark
.sparkContext()
.longAccumulator("old_authors_found");
LongAccumulator updatedAuthorsFoundAcc = spark
.sparkContext()
.longAccumulator("updated_authors_found");
LongAccumulator newAuthorsFoundAcc = spark
.sparkContext()
.longAccumulator("new_authors_found");
LongAccumulator errorCodeAuthorsFoundAcc = spark
.sparkContext()
.longAccumulator("error_code_authors_found");
LongAccumulator errorLoadingAuthorsJsonFoundAcc = spark
.sparkContext()
.longAccumulator("error_loading_authors_json_found");
LongAccumulator errorParsingAuthorsXMLFoundAcc = spark
.sparkContext()
.longAccumulator("error_parsing_authors_xml_found");
Function<Tuple2<Text, Text>, AuthorSummary> retrieveAuthorSummaryFunction = data -> {
AuthorSummary authorSummary = new AuthorSummary();
String orcidId = data._1().toString();
String jsonData = data._2().toString();
JsonElement jElement = new JsonParser().parse(jsonData);
String statusCode = getJsonValue(jElement, "statusCode");
String downloadDate = getJsonValue(jElement, "lastModifiedDate");
if (statusCode.equals("200")) {
String compressedData = getJsonValue(jElement, "compressedData");
if (StringUtils.isEmpty(compressedData)) {
errorLoadingAuthorsJsonFoundAcc.add(1);
} else {
String xmlAuthor = ArgumentApplicationParser.decompressValue(compressedData);
try {
authorSummary = XMLRecordParser
.VTDParseAuthorSummary(xmlAuthor.getBytes());
authorSummary.setStatusCode(statusCode);
authorSummary.setDownloadDate("2020-11-18 00:00:05.644768");
authorSummary.setBase64CompressData(compressedData);
return authorSummary;
} catch (Exception e) {
logger.error("parsing xml " + orcidId + " [" + jsonData + "]", e);
errorParsingAuthorsXMLFoundAcc.add(1);
}
}
} else {
authorSummary.setStatusCode(statusCode);
authorSummary.setDownloadDate("2020-11-18 00:00:05.644768");
errorCodeAuthorsFoundAcc.add(1);
}
return authorSummary;
};
Dataset<AuthorSummary> downloadedAuthorSummaryDS = spark
.createDataset(
sc
.sequenceFile(workingPath + "downloads/updated_authors/*", Text.class, Text.class)
.map(retrieveAuthorSummaryFunction)
.rdd(),
Encoders.bean(AuthorSummary.class));
Dataset<AuthorSummary> currentAuthorSummaryDS = spark
.createDataset(
sc
.textFile(workingPath.concat("orcid_dataset/authors/*"))
.map(item -> OBJECT_MAPPER.readValue(item, AuthorSummary.class))
.rdd(),
Encoders.bean(AuthorSummary.class));
currentAuthorSummaryDS
.joinWith(
downloadedAuthorSummaryDS,
currentAuthorSummaryDS
.col("authorData.oid")
.equalTo(downloadedAuthorSummaryDS.col("authorData.oid")),
"full_outer")
.map(value -> {
Optional<AuthorSummary> opCurrent = Optional.ofNullable(value._1());
Optional<AuthorSummary> opDownloaded = Optional.ofNullable(value._2());
if (!opCurrent.isPresent()) {
newAuthorsFoundAcc.add(1);
return opDownloaded.get();
}
if (!opDownloaded.isPresent()) {
oldAuthorsFoundAcc.add(1);
return opCurrent.get();
}
if (opCurrent.isPresent() && opDownloaded.isPresent()) {
updatedAuthorsFoundAcc.add(1);
return opDownloaded.get();
}
return null;
},
Encoders.bean(AuthorSummary.class))
.filter(Objects::nonNull)
.toJavaRDD()
.map(authorSummary -> OBJECT_MAPPER.writeValueAsString(authorSummary))
.saveAsTextFile(workingPath.concat("orcid_dataset/new_authors"), GzipCodec.class);
logger.info("oldAuthorsFoundAcc: " + oldAuthorsFoundAcc.value().toString());
logger.info("newAuthorsFoundAcc: " + newAuthorsFoundAcc.value().toString());
logger.info("updatedAuthorsFoundAcc: " + updatedAuthorsFoundAcc.value().toString());
logger.info("errorCodeFoundAcc: " + errorCodeAuthorsFoundAcc.value().toString());
logger.info("errorLoadingJsonFoundAcc: " + errorLoadingAuthorsJsonFoundAcc.value().toString());
logger.info("errorParsingXMLFoundAcc: " + errorParsingAuthorsXMLFoundAcc.value().toString());
});
}
private static String getJsonValue(JsonElement jElement, String property) {
if (jElement.getAsJsonObject().has(property)) {
JsonElement name = null;
name = jElement.getAsJsonObject().get(property);
if (name != null && !name.isJsonNull()) {
return name.getAsString();
}
}
return "";
}
}

View File

@ -4,27 +4,23 @@ package eu.dnetlib.doiboost.orcid;
import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkSession;
import java.io.IOException;
import java.util.*;
import java.util.Objects;
import java.util.Optional;
import org.apache.commons.io.IOUtils;
import org.apache.commons.lang3.StringUtils;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.io.compress.GzipCodec;
import org.apache.spark.SparkConf;
import org.apache.spark.api.java.JavaPairRDD;
import org.apache.spark.api.java.JavaRDD;
import org.apache.spark.api.java.JavaSparkContext;
import org.apache.spark.api.java.function.FlatMapFunction;
import org.apache.spark.api.java.function.Function;
import org.apache.spark.api.java.function.MapFunction;
import org.apache.spark.api.java.function.PairFunction;
import org.apache.spark.rdd.RDD;
import org.apache.spark.sql.Dataset;
import org.apache.spark.sql.Encoders;
import org.apache.spark.util.LongAccumulator;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import com.fasterxml.jackson.annotation.JsonInclude;
import com.fasterxml.jackson.databind.ObjectMapper;
import com.google.gson.JsonElement;
import com.google.gson.JsonParser;
@ -33,15 +29,14 @@ import eu.dnetlib.dhp.application.ArgumentApplicationParser;
import eu.dnetlib.dhp.schema.orcid.AuthorSummary;
import eu.dnetlib.dhp.schema.orcid.Work;
import eu.dnetlib.dhp.schema.orcid.WorkDetail;
import eu.dnetlib.dhp.utils.DHPUtils;
import eu.dnetlib.doiboost.orcid.xml.XMLRecordParser;
import eu.dnetlib.doiboost.orcidnodoi.json.JsonWriter;
import eu.dnetlib.doiboost.orcidnodoi.xml.XMLRecordParserNoDoi;
import scala.Tuple2;
public class SparkUpdateOrcidDatasets {
private static final ObjectMapper OBJECT_MAPPER = new ObjectMapper();
private static final ObjectMapper OBJECT_MAPPER = new ObjectMapper()
.setSerializationInclusion(JsonInclude.Include.NON_NULL);
public static void main(String[] args) throws IOException, Exception {
Logger logger = LoggerFactory.getLogger(SparkUpdateOrcidDatasets.class);
@ -67,31 +62,40 @@ public class SparkUpdateOrcidDatasets {
spark -> {
JavaSparkContext sc = JavaSparkContext.fromSparkContext(spark.sparkContext());
LongAccumulator oldAuthorsFoundAcc = spark
.sparkContext()
.longAccumulator("old_authors_found");
LongAccumulator updatedAuthorsFoundAcc = spark
.sparkContext()
.longAccumulator("updated_authors_found");
LongAccumulator newAuthorsFoundAcc = spark
.sparkContext()
.longAccumulator("new_authors_found");
LongAccumulator errorCodeAuthorsFoundAcc = spark
.sparkContext()
.longAccumulator("error_code_authors_found");
LongAccumulator errorLoadingAuthorsJsonFoundAcc = spark
.sparkContext()
.longAccumulator("error_loading_authors_json_found");
LongAccumulator errorLoadingAuthorsXMLFoundAcc = spark
.sparkContext()
.longAccumulator("error_loading_authors_xml_found");
LongAccumulator errorParsingAuthorsXMLFoundAcc = spark
.sparkContext()
.longAccumulator("error_parsing_authors_xml_found");
LongAccumulator oldWorksFoundAcc = spark
.sparkContext()
.longAccumulator("old_works_found");
LongAccumulator updatedWorksFoundAcc = spark
.sparkContext()
.longAccumulator("updated_works_found");
LongAccumulator newWorksFoundAcc = spark
.sparkContext()
.longAccumulator("new_works_found");
LongAccumulator errorCodeWorksFoundAcc = spark
.sparkContext()
.longAccumulator("error_code_works_found");
LongAccumulator errorLoadingWorksJsonFoundAcc = spark
.sparkContext()
.longAccumulator("error_loading_works_json_found");
LongAccumulator errorLoadingWorksXMLFoundAcc = spark
.sparkContext()
.longAccumulator("error_loading_works_xml_found");
LongAccumulator errorParsingWorksXMLFoundAcc = spark
.sparkContext()
.longAccumulator("error_parsing_works_xml_found");
@ -138,25 +142,21 @@ public class SparkUpdateOrcidDatasets {
// errorLoadingAuthorsJsonFoundAcc.add(1);
// } else {
// String xmlAuthor = ArgumentApplicationParser.decompressValue(compressedData);
// if (StringUtils.isEmpty(xmlAuthor)) {
// errorLoadingAuthorsXMLFoundAcc.add(1);
// } else {
// try {
// authorSummary = XMLRecordParser
// .VTDParseAuthorSummary(xmlAuthor.getBytes());
// authorSummary.setStatusCode(statusCode);
// authorSummary.setDownloadDate(downloadDate);
// authorSummary.setBase64CompressData(compressedData);
// return authorSummary;
// } catch (Exception e) {
// logger.error("parsing xml " + orcidId + " [" + jsonData + "]", e);
// errorParsingAuthorsXMLFoundAcc.add(1);
// }
// try {
// authorSummary = XMLRecordParser
// .VTDParseAuthorSummary(xmlAuthor.getBytes());
// authorSummary.setStatusCode(statusCode);
// authorSummary.setDownloadDate("2020-11-18 00:00:05.644768");
// authorSummary.setBase64CompressData(compressedData);
// return authorSummary;
// } catch (Exception e) {
// logger.error("parsing xml " + orcidId + " [" + jsonData + "]", e);
// errorParsingAuthorsXMLFoundAcc.add(1);
// }
// }
// } else {
// authorSummary.setStatusCode(statusCode);
// authorSummary.setDownloadDate(downloadDate);
// authorSummary.setDownloadDate("2020-11-18 00:00:05.644768");
// errorCodeAuthorsFoundAcc.add(1);
// }
// return authorSummary;
@ -187,12 +187,15 @@ public class SparkUpdateOrcidDatasets {
// Optional<AuthorSummary> opCurrent = Optional.ofNullable(value._1());
// Optional<AuthorSummary> opDownloaded = Optional.ofNullable(value._2());
// if (!opCurrent.isPresent()) {
// newAuthorsFoundAcc.add(1);
// return opDownloaded.get();
// }
// if (!opDownloaded.isPresent()) {
// oldAuthorsFoundAcc.add(1);
// return opCurrent.get();
// }
// if (opCurrent.isPresent() && opDownloaded.isPresent()) {
// updatedAuthorsFoundAcc.add(1);
// return opDownloaded.get();
// }
// return null;
@ -200,12 +203,14 @@ public class SparkUpdateOrcidDatasets {
// Encoders.bean(AuthorSummary.class))
// .filter(Objects::nonNull)
// .toJavaRDD()
// .map(authorSummary -> JsonWriter.create(authorSummary))
// .map(authorSummary -> OBJECT_MAPPER.writeValueAsString(authorSummary))
// .saveAsTextFile(workingPath.concat("orcid_dataset/new_authors"), GzipCodec.class);
//
// logger.info("oldAuthorsFoundAcc: " + oldAuthorsFoundAcc.value().toString());
// logger.info("newAuthorsFoundAcc: " + newAuthorsFoundAcc.value().toString());
// logger.info("updatedAuthorsFoundAcc: " + updatedAuthorsFoundAcc.value().toString());
// logger.info("errorCodeFoundAcc: " + errorCodeAuthorsFoundAcc.value().toString());
// logger.info("errorLoadingJsonFoundAcc: " + errorLoadingAuthorsJsonFoundAcc.value().toString());
// logger.info("errorLoadingXMLFoundAcc: " + errorLoadingAuthorsXMLFoundAcc.value().toString());
// logger.info("errorParsingXMLFoundAcc: " + errorParsingAuthorsXMLFoundAcc.value().toString());
Function<String, Work> retrieveWorkFunction = jsonData -> {
@ -214,27 +219,22 @@ public class SparkUpdateOrcidDatasets {
String statusCode = getJsonValue(jElement, "statusCode");
work.setStatusCode(statusCode);
String downloadDate = getJsonValue(jElement, "lastModifiedDate");
work.setDownloadDate(downloadDate);
work.setDownloadDate("2020-11-18 00:00:05.644768");
if (statusCode.equals("200")) {
String compressedData = getJsonValue(jElement, "compressedData");
if (StringUtils.isEmpty(compressedData)) {
errorLoadingWorksJsonFoundAcc.add(1);
} else {
String xmlWork = ArgumentApplicationParser.decompressValue(compressedData);
if (StringUtils.isEmpty(xmlWork)) {
errorLoadingWorksXMLFoundAcc.add(1);
} else {
try {
WorkDetail workDetail = XMLRecordParserNoDoi
.VTDParseWorkData(xmlWork.getBytes());
work.setWorkDetail(workDetail);
work.setBase64CompressData(compressedData);
updatedWorksFoundAcc.add(1);
return work;
} catch (Exception e) {
logger.error("parsing xml [" + jsonData + "]", e);
errorParsingWorksXMLFoundAcc.add(1);
}
try {
WorkDetail workDetail = XMLRecordParserNoDoi
.VTDParseWorkData(xmlWork.getBytes());
work.setWorkDetail(workDetail);
work.setBase64CompressData(compressedData);
return work;
} catch (Exception e) {
logger.error("parsing xml [" + jsonData + "]", e);
errorParsingWorksXMLFoundAcc.add(1);
}
}
} else {
@ -275,12 +275,15 @@ public class SparkUpdateOrcidDatasets {
Optional<Work> opCurrent = Optional.ofNullable(value._1());
Optional<Work> opDownloaded = Optional.ofNullable(value._2());
if (!opCurrent.isPresent()) {
newWorksFoundAcc.add(1);
return opDownloaded.get();
}
if (!opDownloaded.isPresent()) {
oldWorksFoundAcc.add(1);
return opCurrent.get();
}
if (opCurrent.isPresent() && opDownloaded.isPresent()) {
updatedWorksFoundAcc.add(1);
return opDownloaded.get();
}
return null;
@ -288,13 +291,14 @@ public class SparkUpdateOrcidDatasets {
Encoders.bean(Work.class))
.filter(Objects::nonNull)
.toJavaRDD()
.map(work -> JsonWriter.create(work))
.map(work -> OBJECT_MAPPER.writeValueAsString(work))
.saveAsTextFile(workingPath.concat("orcid_dataset/new_works"), GzipCodec.class);
logger.info("oldWorksFoundAcc: " + oldWorksFoundAcc.value().toString());
logger.info("newWorksFoundAcc: " + newWorksFoundAcc.value().toString());
logger.info("updatedWorksFoundAcc: " + updatedWorksFoundAcc.value().toString());
logger.info("errorCodeWorksFoundAcc: " + errorCodeWorksFoundAcc.value().toString());
logger.info("errorLoadingJsonWorksFoundAcc: " + errorLoadingWorksJsonFoundAcc.value().toString());
logger.info("errorLoadingXMLWorksFoundAcc: " + errorLoadingWorksXMLFoundAcc.value().toString());
logger.info("errorParsingXMLWorksFoundAcc: " + errorParsingWorksXMLFoundAcc.value().toString());
});

View File

@ -0,0 +1,181 @@
package eu.dnetlib.doiboost.orcid;
import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkSession;
import java.io.IOException;
import java.util.Objects;
import java.util.Optional;
import org.apache.commons.io.IOUtils;
import org.apache.commons.lang3.StringUtils;
import org.apache.hadoop.io.compress.GzipCodec;
import org.apache.spark.SparkConf;
import org.apache.spark.api.java.JavaSparkContext;
import org.apache.spark.api.java.function.Function;
import org.apache.spark.sql.Dataset;
import org.apache.spark.sql.Encoders;
import org.apache.spark.util.LongAccumulator;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import com.fasterxml.jackson.annotation.JsonInclude;
import com.fasterxml.jackson.databind.ObjectMapper;
import com.google.gson.JsonElement;
import com.google.gson.JsonParser;
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
import eu.dnetlib.dhp.schema.orcid.Work;
import eu.dnetlib.dhp.schema.orcid.WorkDetail;
import eu.dnetlib.doiboost.orcidnodoi.xml.XMLRecordParserNoDoi;
public class SparkUpdateOrcidWorks {
private static final ObjectMapper OBJECT_MAPPER = new ObjectMapper()
.setSerializationInclusion(JsonInclude.Include.NON_NULL);
public static void main(String[] args) throws IOException, Exception {
Logger logger = LoggerFactory.getLogger(SparkUpdateOrcidDatasets.class);
final ArgumentApplicationParser parser = new ArgumentApplicationParser(
IOUtils
.toString(
SparkUpdateOrcidDatasets.class
.getResourceAsStream(
"/eu/dnetlib/dhp/doiboost/download_orcid_data.json")));
parser.parseArgument(args);
Boolean isSparkSessionManaged = Optional
.ofNullable(parser.get("isSparkSessionManaged"))
.map(Boolean::valueOf)
.orElse(Boolean.TRUE);
final String workingPath = parser.get("workingPath");
// final String outputPath = parser.get("outputPath");
SparkConf conf = new SparkConf();
runWithSparkSession(
conf,
isSparkSessionManaged,
spark -> {
JavaSparkContext sc = JavaSparkContext.fromSparkContext(spark.sparkContext());
LongAccumulator oldWorksFoundAcc = spark
.sparkContext()
.longAccumulator("old_works_found");
LongAccumulator updatedWorksFoundAcc = spark
.sparkContext()
.longAccumulator("updated_works_found");
LongAccumulator newWorksFoundAcc = spark
.sparkContext()
.longAccumulator("new_works_found");
LongAccumulator errorCodeWorksFoundAcc = spark
.sparkContext()
.longAccumulator("error_code_works_found");
LongAccumulator errorLoadingWorksJsonFoundAcc = spark
.sparkContext()
.longAccumulator("error_loading_works_json_found");
LongAccumulator errorParsingWorksXMLFoundAcc = spark
.sparkContext()
.longAccumulator("error_parsing_works_xml_found");
Function<String, Work> retrieveWorkFunction = jsonData -> {
Work work = new Work();
JsonElement jElement = new JsonParser().parse(jsonData);
String statusCode = getJsonValue(jElement, "statusCode");
work.setStatusCode(statusCode);
String downloadDate = getJsonValue(jElement, "lastModifiedDate");
work.setDownloadDate("2020-11-18 00:00:05.644768");
if (statusCode.equals("200")) {
String compressedData = getJsonValue(jElement, "compressedData");
if (StringUtils.isEmpty(compressedData)) {
errorLoadingWorksJsonFoundAcc.add(1);
} else {
String xmlWork = ArgumentApplicationParser.decompressValue(compressedData);
try {
WorkDetail workDetail = XMLRecordParserNoDoi
.VTDParseWorkData(xmlWork.getBytes());
work.setWorkDetail(workDetail);
work.setBase64CompressData(compressedData);
return work;
} catch (Exception e) {
logger.error("parsing xml [" + jsonData + "]", e);
errorParsingWorksXMLFoundAcc.add(1);
}
}
} else {
errorCodeWorksFoundAcc.add(1);
}
return work;
};
Dataset<Work> downloadedWorksDS = spark
.createDataset(
sc
.textFile(workingPath + "downloads/updated_works/*")
.map(s -> {
return s.substring(21, s.length() - 1);
})
.map(retrieveWorkFunction)
.rdd(),
Encoders.bean(Work.class));
Dataset<Work> currentWorksDS = spark
.createDataset(
sc
.textFile(workingPath.concat("orcid_dataset/works/*"))
.map(item -> OBJECT_MAPPER.readValue(item, Work.class))
.rdd(),
Encoders.bean(Work.class));
currentWorksDS
.joinWith(
downloadedWorksDS,
currentWorksDS
.col("workDetail.id")
.equalTo(downloadedWorksDS.col("workDetail.id"))
.and(
currentWorksDS
.col("workDetail.oid")
.equalTo(downloadedWorksDS.col("workDetail.oid"))),
"full_outer")
.map(value -> {
Optional<Work> opCurrent = Optional.ofNullable(value._1());
Optional<Work> opDownloaded = Optional.ofNullable(value._2());
if (!opCurrent.isPresent()) {
newWorksFoundAcc.add(1);
return opDownloaded.get();
}
if (!opDownloaded.isPresent()) {
oldWorksFoundAcc.add(1);
return opCurrent.get();
}
if (opCurrent.isPresent() && opDownloaded.isPresent()) {
updatedWorksFoundAcc.add(1);
return opDownloaded.get();
}
return null;
},
Encoders.bean(Work.class))
.filter(Objects::nonNull)
.toJavaRDD()
.map(work -> OBJECT_MAPPER.writeValueAsString(work))
.saveAsTextFile(workingPath.concat("orcid_dataset/new_works"), GzipCodec.class);
logger.info("oldWorksFoundAcc: " + oldWorksFoundAcc.value().toString());
logger.info("newWorksFoundAcc: " + newWorksFoundAcc.value().toString());
logger.info("updatedWorksFoundAcc: " + updatedWorksFoundAcc.value().toString());
logger.info("errorCodeWorksFoundAcc: " + errorCodeWorksFoundAcc.value().toString());
logger.info("errorLoadingJsonWorksFoundAcc: " + errorLoadingWorksJsonFoundAcc.value().toString());
logger.info("errorParsingXMLWorksFoundAcc: " + errorParsingWorksXMLFoundAcc.value().toString());
});
}
private static String getJsonValue(JsonElement jElement, String property) {
if (jElement.getAsJsonObject().has(property)) {
JsonElement name = null;
name = jElement.getAsJsonObject().get(property);
if (name != null && !name.isJsonNull()) {
return name.getAsString();
}
}
return "";
}
}

View File

@ -55,18 +55,54 @@
</configuration>
</global>
<start to="UpdateOrcidDatasets"/>
<start to="promoteOrcidAuthorsDataset"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="UpdateOrcidDatasets">
<action name="ResetWorkingPath">
<fs>
<delete path='${workingPath}/orcid_dataset/new_authors'/>
<delete path='${workingPath}/orcid_dataset/new_works'/>
</fs>
<ok to="UpdateOrcidAuthors"/>
<error to="Kill"/>
</action>
<action name="UpdateOrcidAuthors">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn-cluster</master>
<mode>cluster</mode>
<name>UpdateOrcidDatasets</name>
<class>eu.dnetlib.doiboost.orcid.SparkUpdateOrcidDatasets</class>
<name>UpdateOrcidAuthors</name>
<class>eu.dnetlib.doiboost.orcid.SparkUpdateOrcidAuthors</class>
<jar>dhp-doiboost-${projectVersion}.jar</jar>
<spark-opts>
--conf spark.dynamicAllocation.enabled=true
--conf spark.dynamicAllocation.maxExecutors=${spark2MaxExecutors}
--executor-memory=${sparkExecutorMemory}
--driver-memory=${sparkDriverMemory}
--conf spark.extraListeners=${spark2ExtraListeners}
--conf spark.yarn.historyServer.address=${spark2YarnHistoryServerAddress}
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.sql.queryExecutionListeners=${spark2SqlQueryExecutionListeners}
</spark-opts>
<arg>-w</arg><arg>${workingPath}/</arg>
<arg>-n</arg><arg>${nameNode}</arg>
<arg>-f</arg><arg>-</arg>
<arg>-o</arg><arg>-</arg>
<arg>-t</arg><arg>-</arg>
</spark>
<ok to="UpdateOrcidWorks"/>
<error to="Kill"/>
</action>
<action name="UpdateOrcidWorks">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn-cluster</master>
<mode>cluster</mode>
<name>UpdateOrcidWorks</name>
<class>eu.dnetlib.doiboost.orcid.SparkUpdateOrcidWorks</class>
<jar>dhp-doiboost-${projectVersion}.jar</jar>
<spark-opts>
--conf spark.dynamicAllocation.enabled=true
@ -88,5 +124,40 @@
<error to="Kill"/>
</action>
<action name="promoteOrcidAuthorsDataset">
<distcp xmlns="uri:oozie:distcp-action:0.2">
<prepare>
<delete path="${workingPath}/orcid_dataset/authors"/>
<mkdir path="${workingPath}/orcid_dataset/authors"/>
</prepare>
<arg>${workingPath}/orcid_dataset/new_authors/*</arg>
<arg>${workingPath}/orcid_dataset/authors</arg>
</distcp>
<ok to="promoteOrcidWorksDataset"/>
<error to="Kill"/>
</action>
<action name="promoteOrcidWorksDataset">
<distcp xmlns="uri:oozie:distcp-action:0.2">
<prepare>
<delete path="${workingPath}/orcid_dataset/works"/>
<mkdir path="${workingPath}/orcid_dataset/works"/>
</prepare>
<arg>${workingPath}/orcid_dataset/new_works/*</arg>
<arg>${workingPath}/orcid_dataset/works</arg>
</distcp>
<ok to="CleanWorkingPath"/>
<error to="Kill"/>
</action>
<action name="CleanWorkingPath">
<fs>
<delete path='${workingPath}/orcid_dataset/new_authors'/>
<delete path='${workingPath}/orcid_dataset/new_works'/>
</fs>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>

View File

@ -66,7 +66,7 @@
</configuration>
</global>
<start to="DownloadOrcidWorks"/>
<start to="ResetWorkingPath"/>
<kill name="Kill">
@ -96,21 +96,6 @@
<error to="Kill"/>
</action>
<action name="DownloadUpdatedXMLAuthors">
<java>
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<main-class>eu.dnetlib.doiboost.orcid.OrcidDownloader</main-class>
<arg>-w</arg><arg>${workingPath}/</arg>
<arg>-n</arg><arg>${nameNode}</arg>
<arg>-f</arg><arg>last_modified.csv.tar</arg>
<arg>-o</arg><arg>downloads/</arg>
<arg>-t</arg><arg>${token}</arg>
</java>
<ok to="End"/>
<error to="Kill"/>
</action>
<action name="GenLastModifiedSeq">
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn-cluster</master>