andreas.czerniak
/
BrBETA_dnet-hadoop
forked from D-Net/dnet-hadoop
1
1
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

[cleaning] allow enriched orcids to pass the cleaning, rule out non-orcid author pids

This commit is contained in:
Claudio Atzori 2023-12-12 09:57:00 +01:00
parent aba95ed1d1
commit 84d54643cf
1 changed files with 3 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
dhp-common/src/main/java/eu/dnetlib/dhp/schema/oaf/utils/GraphCleaningFunctions.java
View File

@ -689,6 +689,7 @@ public class GraphCleaningFunctions extends CleaningFunctions {
.filter(Objects::nonNull)
.filter(p -> Objects.nonNull(p.getQualifier()))
.filter(p -> StringUtils.isNotBlank(p.getValue()))
.filter(p -> StringUtils.contains(StringUtils.lowerCase(p.getQualifier().getClassid()), ORCID))
.map(p -> {
// hack to distinguish orcid from orcid_pending
String pidProvenance = getProvenance(p.getDataInfo());
@ -698,7 +699,8 @@ public class GraphCleaningFunctions extends CleaningFunctions {
.toLowerCase()
.contains(ModelConstants.ORCID)) {
if (pidProvenance
.equals(ModelConstants.SYSIMPORT_CROSSWALK_ENTITYREGISTRY)) {
.equals(ModelConstants.SYSIMPORT_CROSSWALK_ENTITYREGISTRY) ||
pidProvenance.equals("ORCID_ENRICHMENT")) {
p.getQualifier().setClassid(ModelConstants.ORCID);
} else {
p.getQualifier().setClassid(ModelConstants.ORCID_PENDING);