129 lines
3.6 KiB
Java
129 lines
3.6 KiB
Java
package prototype;
|
|
|
|
import java.nio.charset.StandardCharsets;
|
|
import java.text.Normalizer;
|
|
import java.util.List;
|
|
import java.util.Set;
|
|
|
|
import prototype.utils.Capitalize;
|
|
import prototype.utils.DotAbbreviations;
|
|
|
|
import com.google.common.base.Joiner;
|
|
import com.google.common.base.Splitter;
|
|
import com.google.common.collect.Iterables;
|
|
import com.google.common.collect.Lists;
|
|
import com.google.common.hash.Hashing;
|
|
|
|
//import eu.dnetlib.pace.clustering.NGramUtils;
|
|
//import eu.dnetlib.pace.util.Capitalise;
|
|
//import eu.dnetlib.pace.util.DotAbbreviations;
|
|
|
|
public class PersonOrig {
|
|
private List<String> name = Lists.newArrayList();
|
|
private List<String> surname = Lists.newArrayList();
|
|
private List<String> fullname = Lists.newArrayList();
|
|
|
|
private static Set<String> particles = null;
|
|
|
|
public PersonOrig(String s) {
|
|
s = Normalizer.normalize(s, Normalizer.Form.NFD);
|
|
s = s.replaceAll("\\(.+\\)", "");
|
|
s = s.replaceAll("\\[.+\\]", "");
|
|
s = s.replaceAll("\\{.+\\}", "");
|
|
s = s.replaceAll("\\s+-\\s+", "-");
|
|
s = s.replaceAll("[\\W&&[^,-]]", " ");
|
|
s = s.replaceAll("[\\p{InCombiningDiacriticalMarks}&&[^,-]]", " ");
|
|
s = s.replaceAll("[\\p{Punct}&&[^-,]]", " ");
|
|
s = s.replaceAll("\\d", " ");
|
|
s = s.replaceAll("\\n", " ");
|
|
s = s.replaceAll("\\.", " ");
|
|
s = s.replaceAll("\\s+", " ");
|
|
|
|
if (s.contains(",")) {
|
|
String[] arr = s.split(",");
|
|
if (arr.length == 1) {
|
|
fullname = splitTerms(arr[0]);
|
|
} else if (arr.length > 1) {
|
|
surname = splitTerms(arr[0]);
|
|
name = splitTerms(arr[1]);
|
|
fullname.addAll(surname);
|
|
fullname.addAll(name);
|
|
}
|
|
} else {
|
|
fullname = splitTerms(s);
|
|
|
|
int lastInitialPosition = fullname.size();
|
|
boolean hasSurnameInUpperCase = false;
|
|
|
|
for (int i = 0; i < fullname.size(); i++) {
|
|
String term = fullname.get(i);
|
|
if (term.length() == 1) {
|
|
lastInitialPosition = i;
|
|
} else if (term.equals(term.toUpperCase())) {
|
|
hasSurnameInUpperCase = true;
|
|
}
|
|
}
|
|
|
|
if (lastInitialPosition < fullname.size() - 1) { // Case: Michele G. Artini
|
|
name = fullname.subList(0, lastInitialPosition + 1);
|
|
surname = fullname.subList(lastInitialPosition + 1, fullname.size());
|
|
} else if (hasSurnameInUpperCase) { // Case: Michele ARTINI
|
|
for (String term : fullname) {
|
|
if (term.length() > 1 && term.equals(term.toUpperCase())) {
|
|
surname.add(term);
|
|
} else {
|
|
name.add(term);
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
private List<String> splitTerms(String s) {
|
|
// if (particles == null) {
|
|
// particles = NGramUtils.loadFromClasspath("/eu/dnetlib/pace/config/name_particles.txt");
|
|
// }
|
|
|
|
List<String> list = Lists.newArrayList();
|
|
for (String part : Splitter.on(" ").omitEmptyStrings().split(s)) {
|
|
// if (!particles.contains(part.toLowerCase())) {
|
|
list.add(part);
|
|
// }
|
|
}
|
|
return list;
|
|
}
|
|
|
|
public List<String> getName() {
|
|
return name;
|
|
}
|
|
|
|
public List<String> getSurname() {
|
|
return surname;
|
|
}
|
|
|
|
public List<String> getFullname() {
|
|
return fullname;
|
|
}
|
|
|
|
public String hash() {
|
|
return Hashing.murmur3_128().hashString(getNormalisedFullname(), StandardCharsets.UTF_8).toString();
|
|
}
|
|
|
|
public String getNormalisedFullname() {
|
|
return isAccurate() ?
|
|
Joiner.on(" ").join(getCapitalSurname()) + ", " + Joiner.on(" ").join(getNameWithAbbreviations()) :
|
|
Joiner.on(" ").join(fullname);
|
|
}
|
|
|
|
public List<String> getCapitalSurname() {
|
|
return Lists.newArrayList(Iterables.transform(surname, new Capitalize()));
|
|
}
|
|
|
|
public List<String> getNameWithAbbreviations() {
|
|
return Lists.newArrayList(Iterables.transform(name, new DotAbbreviations()));
|
|
}
|
|
|
|
public boolean isAccurate() {
|
|
return (name != null && surname != null && !name.isEmpty() && !surname.isEmpty());
|
|
}
|
|
} |