fix dataset creation for downloaded works

This commit is contained in:
Enrico Ottonello 2020-12-11 16:49:54 +01:00
parent 2233750a37
commit 858efbfad1
5 changed files with 288 additions and 39 deletions

View File

@ -3,6 +3,8 @@ package eu.dnetlib.dhp.schema.orcid;
public class OrcidData { public class OrcidData {
protected String base64CompressData; protected String base64CompressData;
protected String statusCode;
protected String downloadDate;
public String getBase64CompressData() { public String getBase64CompressData() {
return base64CompressData; return base64CompressData;
@ -11,4 +13,20 @@ public class OrcidData {
public void setBase64CompressData(String base64CompressData) { public void setBase64CompressData(String base64CompressData) {
this.base64CompressData = base64CompressData; this.base64CompressData = base64CompressData;
} }
public String getStatusCode() {
return statusCode;
}
public void setStatusCode(String statusCode) {
this.statusCode = statusCode;
}
public String getDownloadDate() {
return downloadDate;
}
public void setDownloadDate(String downloadDate) {
this.downloadDate = downloadDate;
}
} }

View File

@ -4,30 +4,47 @@ package eu.dnetlib.doiboost.orcid;
import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkSession; import static eu.dnetlib.dhp.common.SparkSessionSupport.runWithSparkSession;
import java.io.IOException; import java.io.IOException;
import java.util.Optional; import java.util.*;
import org.apache.commons.io.IOUtils; import org.apache.commons.io.IOUtils;
import org.apache.commons.lang3.StringUtils;
import org.apache.hadoop.io.Text; import org.apache.hadoop.io.Text;
import org.apache.hadoop.io.compress.GzipCodec; import org.apache.hadoop.io.compress.GzipCodec;
import org.apache.spark.SparkConf; import org.apache.spark.SparkConf;
import org.apache.spark.api.java.JavaPairRDD; 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.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.Logger;
import org.slf4j.LoggerFactory; import org.slf4j.LoggerFactory;
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.application.ArgumentApplicationParser;
import eu.dnetlib.dhp.schema.orcid.AuthorSummary; import eu.dnetlib.dhp.schema.orcid.AuthorSummary;
import eu.dnetlib.dhp.schema.orcid.Work; import eu.dnetlib.dhp.schema.orcid.Work;
import eu.dnetlib.dhp.schema.orcid.WorkDetail; import eu.dnetlib.dhp.schema.orcid.WorkDetail;
import eu.dnetlib.dhp.utils.DHPUtils;
import eu.dnetlib.doiboost.orcid.xml.XMLRecordParser; import eu.dnetlib.doiboost.orcid.xml.XMLRecordParser;
import eu.dnetlib.doiboost.orcidnodoi.json.JsonWriter; import eu.dnetlib.doiboost.orcidnodoi.json.JsonWriter;
import eu.dnetlib.doiboost.orcidnodoi.xml.XMLRecordParserNoDoi; import eu.dnetlib.doiboost.orcidnodoi.xml.XMLRecordParserNoDoi;
import scala.Tuple2;
public class SparkUpdateOrcidDatasets { public class SparkUpdateOrcidDatasets {
private static final ObjectMapper OBJECT_MAPPER = new ObjectMapper();
public static void main(String[] args) throws IOException, Exception { public static void main(String[] args) throws IOException, Exception {
Logger logger = LoggerFactory.getLogger(SparkUpdateOrcidDatasets.class); Logger logger = LoggerFactory.getLogger(SparkUpdateOrcidDatasets.class);
logger.info("[ SparkUpdateOrcidDatasets STARTED]");
final ArgumentApplicationParser parser = new ArgumentApplicationParser( final ArgumentApplicationParser parser = new ArgumentApplicationParser(
IOUtils IOUtils
@ -40,11 +57,8 @@ public class SparkUpdateOrcidDatasets {
.ofNullable(parser.get("isSparkSessionManaged")) .ofNullable(parser.get("isSparkSessionManaged"))
.map(Boolean::valueOf) .map(Boolean::valueOf)
.orElse(Boolean.TRUE); .orElse(Boolean.TRUE);
logger.info("isSparkSessionManaged: {}", isSparkSessionManaged);
final String workingPath = parser.get("workingPath"); final String workingPath = parser.get("workingPath");
logger.info("workingPath: ", workingPath);
// final String outputPath = parser.get("outputPath"); // final String outputPath = parser.get("outputPath");
// logger.info("outputPath: ", outputPath);
SparkConf conf = new SparkConf(); SparkConf conf = new SparkConf();
runWithSparkSession( runWithSparkSession(
@ -53,35 +67,247 @@ public class SparkUpdateOrcidDatasets {
spark -> { spark -> {
JavaSparkContext sc = JavaSparkContext.fromSparkContext(spark.sparkContext()); JavaSparkContext sc = JavaSparkContext.fromSparkContext(spark.sparkContext());
JavaPairRDD<Text, Text> xmlSummariesRDD = sc LongAccumulator errorCodeAuthorsFoundAcc = spark
.sequenceFile(workingPath.concat("xml/authors/xml_authors.seq"), Text.class, Text.class); .sparkContext()
xmlSummariesRDD .longAccumulator("error_code_authors_found");
.map(seq -> { LongAccumulator errorLoadingAuthorsJsonFoundAcc = spark
AuthorSummary authorSummary = XMLRecordParser .sparkContext()
.VTDParseAuthorSummary(seq._2().toString().getBytes()); .longAccumulator("error_loading_authors_json_found");
authorSummary LongAccumulator errorLoadingAuthorsXMLFoundAcc = spark
.setBase64CompressData(ArgumentApplicationParser.compressArgument(seq._2().toString())); .sparkContext()
return authorSummary; .longAccumulator("error_loading_authors_xml_found");
}) LongAccumulator errorParsingAuthorsXMLFoundAcc = spark
.filter(authorSummary -> authorSummary != null) .sparkContext()
.map(authorSummary -> JsonWriter.create(authorSummary)) .longAccumulator("error_parsing_authors_xml_found");
.saveAsTextFile(workingPath.concat("orcid_dataset/authors"), GzipCodec.class);
JavaPairRDD<Text, Text> xmlWorksRDD = sc LongAccumulator updatedWorksFoundAcc = spark
.sequenceFile(workingPath.concat("xml/works/*"), Text.class, Text.class); .sparkContext()
.longAccumulator("updated_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");
xmlWorksRDD // JavaPairRDD<Text, Text> xmlSummariesRDD = sc
.map(seq -> { // .sequenceFile(workingPath.concat("xml/authors/xml_authors.seq"), Text.class, Text.class);
WorkDetail workDetail = XMLRecordParserNoDoi.VTDParseWorkData(seq._2().toString().getBytes()); // xmlSummariesRDD
Work work = new Work(); // .map(seq -> {
work.setWorkDetail(workDetail); // AuthorSummary authorSummary = XMLRecordParser
work.setBase64CompressData(ArgumentApplicationParser.compressArgument(seq._2().toString())); // .VTDParseAuthorSummary(seq._2().toString().getBytes());
return work; // authorSummary
}) // .setBase64CompressData(ArgumentApplicationParser.compressArgument(seq._2().toString()));
.filter(work -> work != null) // return authorSummary;
// })
// .filter(authorSummary -> authorSummary != null)
// .map(authorSummary -> JsonWriter.create(authorSummary))
// .saveAsTextFile(workingPath.concat("orcid_dataset/authors"), GzipCodec.class);
//
// JavaPairRDD<Text, Text> xmlWorksRDD = sc
// .sequenceFile(workingPath.concat("xml/works/*"), Text.class, Text.class);
//
// xmlWorksRDD
// .map(seq -> {
// WorkDetail workDetail = XMLRecordParserNoDoi.VTDParseWorkData(seq._2().toString().getBytes());
// Work work = new Work();
// work.setWorkDetail(workDetail);
// work.setBase64CompressData(ArgumentApplicationParser.compressArgument(seq._2().toString()));
// return work;
// })
// .filter(work -> work != null)
// .map(work -> JsonWriter.create(work))
// .saveAsTextFile(workingPath.concat("orcid_dataset/works"), GzipCodec.class);
// 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);
// 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);
// }
// }
// }
// } else {
// authorSummary.setStatusCode(statusCode);
// authorSummary.setDownloadDate(downloadDate);
// 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()) {
// return opDownloaded.get();
// }
// if (!opDownloaded.isPresent()) {
// return opCurrent.get();
// }
// if (opCurrent.isPresent() && opDownloaded.isPresent()) {
// return opDownloaded.get();
// }
// return null;
// },
// Encoders.bean(AuthorSummary.class))
// .filter(Objects::nonNull)
// .toJavaRDD()
// .map(authorSummary -> JsonWriter.create(authorSummary))
// .saveAsTextFile(workingPath.concat("orcid_dataset/new_authors"), GzipCodec.class);
//
// 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 -> {
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(downloadDate);
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);
}
}
}
} 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()) {
return opDownloaded.get();
}
if (!opDownloaded.isPresent()) {
return opCurrent.get();
}
if (opCurrent.isPresent() && opDownloaded.isPresent()) {
return opDownloaded.get();
}
return null;
},
Encoders.bean(Work.class))
.filter(Objects::nonNull)
.toJavaRDD()
.map(work -> JsonWriter.create(work)) .map(work -> JsonWriter.create(work))
.saveAsTextFile(workingPath.concat("orcid_dataset/works"), GzipCodec.class); .saveAsTextFile(workingPath.concat("orcid_dataset/new_works"), GzipCodec.class);
});
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());
});
}
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

@ -2,7 +2,7 @@
<parameters> <parameters>
<property> <property>
<name>spark2MaxExecutors</name> <name>spark2MaxExecutors</name>
<value>40</value> <value>50</value>
</property> </property>
<property> <property>
<name>sparkDriverMemory</name> <name>sparkDriverMemory</name>

View File

@ -372,12 +372,17 @@ public class OrcidClientTest {
} }
@Test @Test
public void testReadDownloadedWork() throws Exception { public void testUpdatedRecord() throws Exception {
final String base64CompressedRecord = IOUtils final String base64CompressedRecord = IOUtils
.toString(getClass().getResourceAsStream("0000-0002-6664-7451_work.compressed.base64")); .toString(getClass().getResourceAsStream("0000-0003-3028-6161.compressed.base64"));
final String work = ArgumentApplicationParser.decompressValue(base64CompressedRecord); final String record = ArgumentApplicationParser.decompressValue(base64CompressedRecord);
logToFile("\n\ndownloaded \n\n" + work); logToFile("\n\nrecord updated \n\n" + record);
// final String downloadedRecord = testDownloadRecord("0000-0003-3028-6161", REQUEST_TYPE_RECORD); }
// assertTrue(recordFromSeqFile.equals(downloadedRecord));
@Test
public void testUpdatedWork() throws Exception {
final String base64CompressedWork = "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";
final String work = ArgumentApplicationParser.decompressValue(base64CompressedWork);
logToFile("\n\nwork updated \n\n" + work);
} }
} }