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

removed duplicated subjects

This commit is contained in:
Enrico Ottonello 2022-06-28 13:07:53 +02:00
parent 33dcf9964c
commit 18c9b95cb1
1 changed files with 12 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
dhp-workflows/dhp-rdfconverter/src/main/java/eu/dnetlib/dhp/rdfconverter/utils/RDFConverter.java
View File

@ -254,6 +254,18 @@ public class RDFConverter {
});
});
dataciteProteins
.stream()
.filter(d -> (d.getSubjects() != null && !d.getSubjects().isEmpty()))
.forEach(d -> {
Map<String, DataciteProtein.Subject> cleanedSubjects = new HashMap<>();
d.getSubjects().stream().forEach(s -> {
cleanedSubjects.putIfAbsent(s.getValue(), s);
});
d.getSubjects().clear();
d.setSubjects(cleanedSubjects.values().stream().collect(Collectors.toList()));
});
dataciteProteins.forEach(d -> {
DataciteProtein.Title title = new DataciteProtein.Title();
title.setTitle(d.getId() + " - ".concat(mainTitles.get(0)));