D-Net
/
dnet-hadoop
14
5
Fork 11
Code Issues 14 Pull Requests 13 Projects 2 Releases 2 Wiki Activity

SimpleDateFormat is not thread safe; improved error reporting in case of invalid dates

This commit is contained in:
Claudio Atzori 2020-11-26 11:03:12 +01:00
parent c1b9a4045a
commit 76363a8512
1 changed files with 23 additions and 18 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

41
dhp-workflows/dhp-dedup-openaire/src/main/java/eu/dnetlib/dhp/oa/dedup/model/Identifier.java
View File

@ -4,10 +4,7 @@ package eu.dnetlib.dhp.oa.dedup.model;
import java.io.Serializable; import java.io.Serializable;
import java.text.ParseException; import java.text.ParseException;
import java.text.SimpleDateFormat; import java.text.SimpleDateFormat;
import java.util.Date; import java.util.*;
import java.util.List;
import java.util.Optional;
import java.util.Set;
import java.util.stream.Collectors; import java.util.stream.Collectors;
import org.apache.commons.lang3.StringUtils; import org.apache.commons.lang3.StringUtils;
@ -23,14 +20,16 @@ import eu.dnetlib.dhp.schema.oaf.utils.PidType;
public class Identifier<T extends OafEntity> implements Serializable, Comparable<Identifier> { public class Identifier<T extends OafEntity> implements Serializable, Comparable<Identifier> {
public static final String DATE_FORMAT = "yyyy-MM-dd";
public static String CROSSREF_ID = "10|openaire____::081b82f96300b6a6e3d282bad31cb6e2"; public static String CROSSREF_ID = "10|openaire____::081b82f96300b6a6e3d282bad31cb6e2";
public static String DATACITE_ID = "10|openaire____::9e3be59865b2c1c335d32dae2fe7b254"; public static String DATACITE_ID = "10|openaire____::9e3be59865b2c1c335d32dae2fe7b254";
public static String BASE_DATE = "2000-01-01"; public static String BASE_DATE = "2000-01-01";
private static SimpleDateFormat sdf = new SimpleDateFormat("yyyy-MM-dd");
private T entity; private T entity;
// cached date value
private Date date = null;
public static <T extends OafEntity> Identifier newInstance(T entity) { public static <T extends OafEntity> Identifier newInstance(T entity) {
return new Identifier(entity); return new Identifier(entity);
} }
@ -48,17 +47,23 @@ public class Identifier<T extends OafEntity> implements Serializable, Comparable
} }
public Date getDate() { public Date getDate() {
String date = BASE_DATE; if (Objects.nonNull(date)) {
if (ModelSupport.isSubClass(getEntity(), Result.class)) { return date;
Result result = (Result) getEntity(); } else {
if (isWellformed(result.getDateofacceptance())) { String sDate = BASE_DATE;
date = result.getDateofacceptance().getValue(); if (ModelSupport.isSubClass(getEntity(), Result.class)) {
Result result = (Result) getEntity();
if (isWellformed(result.getDateofacceptance())) {
sDate = result.getDateofacceptance().getValue();
}
}
try {
this.date = new SimpleDateFormat(DATE_FORMAT).parse(sDate);
return date;
} catch (Throwable e) {
throw new RuntimeException(
String.format("cannot parse date: '%s' from record: '%s'", sDate, entity.getId()));
} }
}
try {
return sdf.parse(date);
} catch (ParseException e) {
return new Date();
} }
} }
@ -117,10 +122,10 @@ public class Identifier<T extends OafEntity> implements Serializable, Comparable
} }
if (this.getDate().compareTo(i.getDate()) == 0) {// same date if (this.getDate().compareTo(i.getDate()) == 0) {// same date
// the minus because we need to take the alphabetically lower id // we need to take the alphabetically lower id
return this.getOriginalID().compareTo(i.getOriginalID()); return this.getOriginalID().compareTo(i.getOriginalID());
} else } else
// the minus is because we need to take the elder date // we need to take the elder date
return this.getDate().compareTo(i.getDate()); return this.getDate().compareTo(i.getDate());
} else { } else {
return new PidComparator<>(getEntity()).compare(toSP(getPidType()), toSP(i.getPidType())); return new PidComparator<>(getEntity()).compare(toSP(getPidType()), toSP(i.getPidType()));