forked from D-Net/dnet-hadoop
This commit is contained in:
parent
d08dca0745
commit
33c27b6f75
|
@ -424,6 +424,9 @@ public class MakeReportSparkJob implements Serializable {
|
|||
return new Tuple2<>("check", reportInfo);
|
||||
}
|
||||
|
||||
}else{
|
||||
reportInfo.setLevel("not found a match in name matching");
|
||||
return new Tuple2<>("check", reportInfo);
|
||||
}
|
||||
}
|
||||
|
||||
|
|
|
@ -12,6 +12,7 @@ import java.util.stream.Collectors;
|
|||
import org.apache.commons.io.FileUtils;
|
||||
import org.apache.commons.lang3.StringUtils;
|
||||
import org.apache.commons.text.similarity.CosineDistance;
|
||||
import org.apache.commons.text.similarity.FuzzyScore;
|
||||
import org.apache.hadoop.io.Text;
|
||||
import org.apache.hadoop.security.WhitelistBasedResolver;
|
||||
import org.apache.neethi.Assertion;
|
||||
|
@ -160,9 +161,16 @@ public class CleanOrcidTest {
|
|||
"21", MakeReportSparkJob.handleNameSurname("Matthew Jones")
|
||||
},
|
||||
{
|
||||
"20", MakeReportSparkJob.handleNameSurname("Martin Fenner")
|
||||
"22", MakeReportSparkJob.handleNameSurname("Martin Fenner")
|
||||
},
|
||||
{
|
||||
"23", MakeReportSparkJob.handleNameSurname("Surachai Karnjanakom")
|
||||
},
|
||||
{
|
||||
"24", MakeReportSparkJob.handleNameSurname("Panya Maneechakr")
|
||||
}
|
||||
|
||||
|
||||
};
|
||||
|
||||
@BeforeAll
|
||||
|
@ -544,6 +552,18 @@ public class CleanOrcidTest {
|
|||
});
|
||||
}
|
||||
|
||||
@Test
|
||||
public void testAuthorFuzzyApache(){
|
||||
for (int i = 0; i < input.length; i += 2) {
|
||||
System.out
|
||||
.println(
|
||||
"FuzzyScore of '" + input[i][1] + "' & '" + input[i + 1][1] + "' | Similarity ratio "
|
||||
+ new FuzzyScore(Locale.getDefault()).fuzzyScore(input[i][1], input[i + 1][1]));
|
||||
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
@Test
|
||||
public void FuzzyWuzzyTest() {
|
||||
applyFuzzyWuzzy(input);
|
||||
|
|
Loading…
Reference in New Issue