D-Net
/
dnet-hadoop
14
5
Fork 11
Code Issues 14 Pull Requests 13 Projects 2 Releases 2 Wiki Activity

Merge pull request 'Remove ORCID information when the same ORCID ID is used multiple times in the same result for different authors' (#503) from clean_clashing_orcids into beta 

Reviewed-on: #503
This commit is contained in:
Giambattista Bloisi 2024-11-08 09:31:11 +01:00
parent f7bb53fe78 8f5171557e
commit f31f22801f
1 changed files with 35 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

36
dhp-common/src/main/java/eu/dnetlib/dhp/schema/oaf/utils/GraphCleaningFunctions.java
View File

@ -2,7 +2,6 @@
package eu.dnetlib.dhp.schema.oaf.utils;
import static eu.dnetlib.dhp.schema.common.ModelConstants.*;
import static eu.dnetlib.dhp.schema.common.ModelConstants.OPENAIRE_META_RESOURCE_TYPE;
import static eu.dnetlib.dhp.schema.oaf.utils.OafMapperUtils.getProvenance;
import java.net.MalformedURLException;
@ -696,6 +695,7 @@ public class GraphCleaningFunctions extends CleaningFunctions {
}
}
// set ORCID_PENDING to all orcid values that are not coming from ORCID provenance
for (Author a : r.getAuthor()) {
if (Objects.isNull(a.getPid())) {
a.setPid(Lists.newArrayList());
@ -752,6 +752,40 @@ public class GraphCleaningFunctions extends CleaningFunctions {
.collect(Collectors.toList()));
}
}
// Identify clashing ORCIDS:that is same ORCID associated to multiple authors in this result
Map<String, Integer> clashing_orcid = new HashMap<>();
for (Author a : r.getAuthor()) {
a
.getPid()
.stream()
.filter(
p -> StringUtils
.contains(StringUtils.lowerCase(p.getQualifier().getClassid()), ORCID_PENDING))
.map(StructuredProperty::getValue)
.distinct()
.forEach(orcid -> clashing_orcid.compute(orcid, (k, v) -> (v == null) ? 1 : v + 1));
}
Set<String> clashing = clashing_orcid
.entrySet()
.stream()
.filter(ee -> ee.getValue() > 1)
.map(Map.Entry::getKey)
.collect(Collectors.toSet());
// filter out clashing orcids
for (Author a : r.getAuthor()) {
a
.setPid(
a
.getPid()
.stream()
.filter(p -> !clashing.contains(p.getValue()))
.collect(Collectors.toList()));
}
}
if (value instanceof Publication) {