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

PID filtering based on a blacklist

This commit is contained in:
Claudio Atzori 2020-11-17 12:27:06 +01:00
parent 628ca54dd3
commit cfc01f136e
1 changed files with 10 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
dhp-workflows/dhp-graph-mapper/src/main/java/eu/dnetlib/dhp/oa/graph/clean/CleaningFunctions.java
View File

@ -1,8 +1,10 @@
package eu.dnetlib.dhp.oa.graph.clean;
import java.util.HashSet;
import java.util.LinkedHashMap;
import java.util.Objects;
import java.util.Set;
import java.util.function.Function;
import java.util.stream.Collectors;
@ -17,7 +19,13 @@ import eu.dnetlib.dhp.schema.oaf.*;
public class CleaningFunctions {
public static final String ORCID_PREFIX_REGEX = "^http(s?):\\/\\/orcid\\.org\\/";
public static final String NONE = "none";
public static final Set<String> PID_BLACKLIST = new HashSet<>();
static {
PID_BLACKLIST.add("none");
PID_BLACKLIST.add("na");
}
public static <T extends Oaf> T fixVocabularyNames(T value) {
if (value instanceof Datasource) {
@ -114,7 +122,7 @@ public class CleaningFunctions {
.stream()
.filter(Objects::nonNull)
.filter(sp -> StringUtils.isNotBlank(StringUtils.trim(sp.getValue())))
.filter(sp -> !NONE.equalsIgnoreCase(sp.getValue().trim()))
.filter(sp -> !PID_BLACKLIST.contains(sp.getValue().trim().toLowerCase()))
.filter(sp -> Objects.nonNull(sp.getQualifier()))
.filter(sp -> StringUtils.isNotBlank(sp.getQualifier().getClassid()))
.map(sp -> {