enrico.ottonello
/
dnet-hadoop
forked from D-Net/dnet-hadoop
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

extended existing codo to accomodate import of POCI from open citation

This commit is contained in:
Miriam Baglioni 2023-10-02 09:25:16 +02:00
parent 0935d7757c
commit e84f5b5e64
7 changed files with 67 additions and 23 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

29
dhp-workflows/dhp-aggregation/src/main/java/eu/dnetlib/dhp/actionmanager/opencitations/CreateActionSetSparkJob.java
View File

@ -26,7 +26,6 @@ import eu.dnetlib.dhp.actionmanager.opencitations.model.COCI;
import eu.dnetlib.dhp.application.ArgumentApplicationParser;
import eu.dnetlib.dhp.schema.action.AtomicAction;
import eu.dnetlib.dhp.schema.common.ModelConstants;
import eu.dnetlib.dhp.schema.common.ModelSupport;
import eu.dnetlib.dhp.schema.oaf.*;
import eu.dnetlib.dhp.schema.oaf.utils.CleaningFunctions;
import eu.dnetlib.dhp.schema.oaf.utils.IdentifierFactory;
@ -35,7 +34,9 @@ import scala.Tuple2;
public class CreateActionSetSparkJob implements Serializable {
public static final String OPENCITATIONS_CLASSID = "sysimport:crosswalk:opencitations";
public static final String OPENCITATIONS_CLASSNAME = "Imported from OpenCitations";
private static final String ID_PREFIX = "50|doi_________::";
private static final String DOI_PREFIX = "50|doi_________::";
private static final String PMID_PREFIX = "50|pmid________::";
private static final String TRUST = "0.91";
private static final Logger log = LoggerFactory.getLogger(CreateActionSetSparkJob.class);
@ -67,6 +68,9 @@ public class CreateActionSetSparkJob implements Serializable {
final String outputPath = parser.get("outputPath");
log.info("outputPath {}", outputPath);
final String prefix = parser.get("prefix");
log.info("prefix {}", prefix);
final boolean shouldDuplicateRels = Optional
.ofNullable(parser.get("shouldDuplicateRels"))
.map(Boolean::valueOf)
@ -77,13 +81,13 @@ public class CreateActionSetSparkJob implements Serializable {
conf,
isSparkSessionManaged,
spark -> {
extractContent(spark, inputPath, outputPath, shouldDuplicateRels);
extractContent(spark, inputPath, outputPath, shouldDuplicateRels, prefix);
});
}
private static void extractContent(SparkSession spark, String inputPath, String outputPath,
boolean shouldDuplicateRels) {
boolean shouldDuplicateRels, String prefix) {
spark
.read()
.textFile(inputPath + "/*")
@ -91,7 +95,8 @@ public class CreateActionSetSparkJob implements Serializable {
(MapFunction<String, COCI>) value -> OBJECT_MAPPER.readValue(value, COCI.class),
Encoders.bean(COCI.class))
.flatMap(
(FlatMapFunction<COCI, Relation>) value -> createRelation(value, shouldDuplicateRels).iterator(),
(FlatMapFunction<COCI, Relation>) value -> createRelation(value, shouldDuplicateRels, prefix)
.iterator(),
Encoders.bean(Relation.class))
.filter((FilterFunction<Relation>) value -> value != null)
.toJavaRDD()
@ -103,13 +108,19 @@ public class CreateActionSetSparkJob implements Serializable {
}
private static List<Relation> createRelation(COCI value, boolean duplicate) {
private static List<Relation> createRelation(COCI value, boolean duplicate, String p) {
List<Relation> relationList = new ArrayList<>();
String prefix;
if (p.equals("COCI")) {
prefix = DOI_PREFIX;
} else {
prefix = PMID_PREFIX;
}
String citing = ID_PREFIX
String citing = prefix
+ IdentifierFactory.md5(CleaningFunctions.normalizePidValue("doi", value.getCiting()));
final String cited = ID_PREFIX
final String cited = prefix
+ IdentifierFactory.md5(CleaningFunctions.normalizePidValue("doi", value.getCited()));
if (!citing.equals(cited)) {
@ -120,7 +131,7 @@ public class CreateActionSetSparkJob implements Serializable {
cited, ModelConstants.CITES));
if (duplicate && value.getCiting().endsWith(".refs")) {
citing = ID_PREFIX + IdentifierFactory
citing = prefix + IdentifierFactory
.md5(
CleaningFunctions
.normalizePidValue(

16
dhp-workflows/dhp-aggregation/src/main/java/eu/dnetlib/dhp/actionmanager/opencitations/GetOpenCitationsRefs.java
View File

@ -45,6 +45,9 @@ public class GetOpenCitationsRefs implements Serializable {
final String hdfsNameNode = parser.get("hdfsNameNode");
log.info("hdfsNameNode {}", hdfsNameNode);
final String prefix = parser.get("prefix");
log.info("prefix {}", prefix);
Configuration conf = new Configuration();
conf.set("fs.defaultFS", hdfsNameNode);
@ -53,30 +56,31 @@ public class GetOpenCitationsRefs implements Serializable {
GetOpenCitationsRefs ocr = new GetOpenCitationsRefs();
for (String file : inputFile) {
ocr.doExtract(workingPath + "/Original/" + file, workingPath, fileSystem);
ocr.doExtract(workingPath + "/Original/" + file, workingPath, fileSystem, prefix);
}
}
private void doExtract(String inputFile, String workingPath, FileSystem fileSystem)
private void doExtract(String inputFile, String workingPath, FileSystem fileSystem, String prefix)
throws IOException {
final Path path = new Path(inputFile);
FSDataInputStream oc_zip = fileSystem.open(path);
int count = 1;
// int count = 1;
try (ZipInputStream zis = new ZipInputStream(oc_zip)) {
ZipEntry entry = null;
while ((entry = zis.getNextEntry()) != null) {
if (!entry.isDirectory()) {
String fileName = entry.getName();
fileName = fileName.substring(0, fileName.indexOf("T")) + "_" + count;
count++;
// fileName = fileName.substring(0, fileName.indexOf("T")) + "_" + count;
fileName = fileName.substring(0, fileName.lastIndexOf("."));
// count++;
try (
FSDataOutputStream out = fileSystem
.create(new Path(workingPath + "/COCI/" + fileName + ".gz"));
.create(new Path(workingPath + "/" + prefix + "/" + fileName + ".gz"));
GZIPOutputStream gzipOs = new GZIPOutputStream(new BufferedOutputStream(out))) {
IOUtils.copy(zis, gzipOs);

18
dhp-workflows/dhp-aggregation/src/main/java/eu/dnetlib/dhp/actionmanager/opencitations/ReadCOCI.java
View File

@ -49,6 +49,9 @@ public class ReadCOCI implements Serializable {
final String workingPath = parser.get("workingPath");
log.info("workingPath {}", workingPath);
final String format = parser.get("format");
log.info("format {}", format);
SparkConf sconf = new SparkConf();
final String delimiter = Optional
@ -64,13 +67,14 @@ public class ReadCOCI implements Serializable {
workingPath,
inputFile,
outputPath,
delimiter);
delimiter,
format);
});
}
private static void doRead(SparkSession spark, String workingPath, String[] inputFiles,
String outputPath,
String delimiter) throws IOException {
String delimiter, String format) throws IOException {
for (String inputFile : inputFiles) {
String p_string = workingPath + "/" + inputFile + ".gz";
@ -87,9 +91,15 @@ public class ReadCOCI implements Serializable {
cociData.map((MapFunction<Row, COCI>) row -> {
COCI coci = new COCI();
if (format.equals("COCI")) {
coci.setCiting(row.getString(1));
coci.setCited(row.getString(2));
} else {
coci.setCiting(String.valueOf(row.getInt(1)));
coci.setCited(String.valueOf(row.getInt(2)));
}
coci.setOci(row.getString(0));
coci.setCiting(row.getString(1));
coci.setCited(row.getString(2));
return coci;
}, Encoders.bean(COCI.class))
.write()

5
dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/actionmanager/opencitations/as_parameters.json
View File

@ -21,5 +21,10 @@
"paramLongName": "shouldDuplicateRels",
"paramDescription": "the hdfs name node",
"paramRequired": false
},{
"paramName": "p",
"paramLongName": "prefix",
"paramDescription": "the hdfs name node",
"paramRequired": true
}
]

6
dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/actionmanager/opencitations/input_parameters.json
View File

@ -16,5 +16,11 @@
"paramLongName": "hdfsNameNode",
"paramDescription": "the hdfs name node",
"paramRequired": true
},
{
"paramName": "p",
"paramLongName": "prefix",
"paramDescription": "COCI or POCI",
"paramRequired": true
}
]

7
dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/actionmanager/opencitations/input_readcoci_parameters.json
View File

@ -30,7 +30,12 @@
"paramLongName": "inputFile",
"paramDescription": "the hdfs name node",
"paramRequired": true
}
}, {
"paramName": "f",
"paramLongName": "format",
"paramDescription": "the hdfs name node",
"paramRequired": true
}
]

9
dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/actionmanager/opencitations/oozie_app/workflow.xml
View File

@ -60,6 +60,7 @@
<arg>--hdfsNameNode</arg><arg>${nameNode}</arg>
<arg>--inputFile</arg><arg>${inputFile}</arg>
<arg>--workingPath</arg><arg>${workingPath}</arg>
<arg>--prefix</arg><arg>${prefix}</arg>
</java>
<ok to="read"/>
<error to="Kill"/>
@ -82,10 +83,11 @@
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
</spark-opts>
<arg>--workingPath</arg><arg>${workingPath}/COCI</arg>
<arg>--outputPath</arg><arg>${workingPath}/COCI_JSON/</arg>
<arg>--workingPath</arg><arg>${workingPath}/${prefix}</arg>
<arg>--outputPath</arg><arg>${workingPath}/${prefix}_JSON/</arg>
<arg>--delimiter</arg><arg>${delimiter}</arg>
<arg>--inputFile</arg><arg>${inputFileCoci}</arg>
<arg>--format</arg><arg>${prefix}</arg>
</spark>
<ok to="create_actionset"/>
<error to="Kill"/>
@ -108,8 +110,9 @@
--conf spark.eventLog.dir=${nameNode}${spark2EventLogDir}
--conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
</spark-opts>
<arg>--inputPath</arg><arg>${workingPath}/COCI_JSON</arg>
<arg>--inputPath</arg><arg>${workingPath}/${prefix}_JSON</arg>
<arg>--outputPath</arg><arg>${outputPath}</arg>
<arg>--prefix</arg><arg>${prefix}</arg>
</spark>
<ok to="End"/>
<error to="Kill"/>