michele.artini
/
dnet-hadoop
forked from D-Net/dnet-hadoop
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

fix bug on oai iterator that skip record cleaned

This commit is contained in:
Sandro La Bruzzo 2021-09-07 11:20:41 +02:00
parent d64a942a76
commit 370dddb2fa
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
dhp-workflows/dhp-aggregation/src/main/java/eu/dnetlib/dhp/collection/plugin/oai/OaiIterator.java
View File

@ -161,7 +161,7 @@ public class OaiIterator implements Iterator<String> {
report.put(e.getClass().getName(), e.getMessage());
final String cleaned = XmlCleaner.cleanAllEntities(xml);
try {
doc = DocumentHelper.parseText(xml);
doc = DocumentHelper.parseText(cleaned);
} catch (final DocumentException e1) {
final String resumptionToken = extractResumptionToken(xml);
if (resumptionToken == null) {