D-Net
/
dnet-hadoop
15
4
Fork 10
Code Issues 14 Pull Requests 8 Projects 2 Releases 1 Wiki Activity

test fr TextGrid

Browse Source
pull/60/head
Alessia Bardi 3 years ago
parent cfc01f136e
commit 7e0a76a8ac
2 changed files with 129 additions and 0 deletions
Show Stats Download Patch File Download Diff File

16
dhp-workflows/dhp-graph-mapper/src/test/java/eu/dnetlib/dhp/oa/graph/raw/MappersTest.java
Unescape Escape View File

@ -340,6 +340,22 @@ public class MappersTest {
assertTrue(StringUtils.isNotBlank(p.getTitle().get(0).getValue()));
}
@Test
void testTextGrid() throws IOException {
final String xml = IOUtils.toString(getClass().getResourceAsStream("textgrid.xml"));
final List<Oaf> list = new OdfToOafMapper(vocs, false).processMdRecord(xml);
System.out.println("***************");
System.out.println(new ObjectMapper().writeValueAsString(list));
System.out.println("***************");
final Dataset p = (Dataset) list.get(0);
assertValidId(p.getId());
assertValidId(p.getCollectedfrom().get(0).getKey());
assertTrue(StringUtils.isNotBlank(p.getTitle().get(0).getValue()));
System.out.println(p.getTitle().get(0).getValue());
}
private void assertValidId(final String id) {
assertEquals(49, id.length());
assertEquals('|', id.charAt(2));

113
dhp-workflows/dhp-graph-mapper/src/test/resources/eu/dnetlib/dhp/oa/graph/raw/textgrid.xml
Unescape Escape View File

@ -0,0 +1,113 @@
<?xml version="1.0" encoding="UTF-8"?>
<record xmlns:dr="http://www.driver-repository.eu/namespace/dr"
xmlns:oaf="http://namespace.openaire.eu/oaf"
xmlns:oai="http://www.openarchives.org/OAI/2.0/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
<oai:header xmlns="http://namespace.openaire.eu/"
xmlns:dc="http://purl.org/dc/elements/1.1/"
xmlns:dri="http://www.driver-repository.eu/namespace/dri" xmlns:prov="http://www.openarchives.org/OAI/2.0/provenance">
<dri:objIdentifier>r3f52792889d::000051aa1f61d77d2c0b340091f8024e</dri:objIdentifier>
<dri:recordIdentifier>textgrid:q9cv.0</dri:recordIdentifier>
<dri:dateOfCollection>2020-11-17T09:34:11.128+01:00</dri:dateOfCollection>
<oaf:datasourceprefix>r3f52792889d</oaf:datasourceprefix>
<identifier xmlns="http://www.openarchives.org/OAI/2.0/">textgrid:q9cv.0</identifier>
<datestamp xmlns="http://www.openarchives.org/OAI/2.0/">2012-01-21T13:35:20Z</datestamp>
<dr:dateOfTransformation>2020-11-17T09:46:21.551+01:00</dr:dateOfTransformation>
</oai:header>
<metadata>
<datacite:resource xmlns="http://www.openarchives.org/OAI/2.0/"
xmlns:datacite="http://datacite.org/schema/kernel-3"
xmlns:dc="http://purl.org/dc/elements/1.1/"
xmlns:dri="http://www.driver-repository.eu/namespace/dri" xmlns:prov="http://www.openarchives.org/OAI/2.0/provenance">
<datacite:identifier identifierType="Handle">hdl:11858/00-1734-0000-0003-7664-F</datacite:identifier>
<datacite:creators>
<datacite:creator>
<datacite:creatorName>Hoffmann von Fallersleben, August Heinrich</datacite:creatorName>
<datacite:nameIdentifier nameIdentifierScheme="pnd" schemeURI="https://de.dariah.eu/pnd-service">118552589</datacite:nameIdentifier>
</datacite:creator>
</datacite:creators>
<datacite:titles>
<datacite:title titleType="Other">Mailied</datacite:title>
<datacite:title titleType="Other">August Heinrich Hoffmann von Fallersleben: Unpolitische Lieder von Hoffmann von Fallersleben, 1. + 2. Theil, 1. Theil, Hamburg: Hoffmann und Campe, 1841.</datacite:title>
</datacite:titles>
<datacite:publisher>TextGrid</datacite:publisher>
<datacite:publicationYear>2012</datacite:publicationYear>
<datacite:contributors>
<datacite:contributor contributorType="DataManager">
<datacite:contributorName>tvitt@textgrid.de</datacite:contributorName>
</datacite:contributor>
<datacite:contributor contributorType="Other">
<datacite:contributorName>Digitale Bibliothek</datacite:contributorName>
<datacite:nameIdentifier nameIdentifierScheme="textgrid">TGPR-372fe6dc-57f2-6cd4-01b5-2c4bbefcfd3c</datacite:nameIdentifier>
</datacite:contributor>
</datacite:contributors>
<datacite:dates>
<datacite:date dateType="Created">2012-01-21T13:35:20Z</datacite:date>
<datacite:date dateType="Issued">2012-01-21T13:35:20Z</datacite:date>
<datacite:date dateType="Updated">2012-01-21T13:35:20Z</datacite:date>
</datacite:dates>
<datacite:resourceType resourceTypeGeneral="Dataset"/>
<alternateIdentifiers>
<datacite:alternateIdentifier alternateIdentifierType="URI">textgrid:q9cv.0</datacite:alternateIdentifier>
<alternateIdentifier alternateIdentifierType="URL">http://hdl.handle.net/hdl:11858/00-1734-0000-0003-7664-F</alternateIdentifier>
</alternateIdentifiers>
<datacite:relatedIdentifiers>
<datacite:relatedIdentifier relatedIdentifierType="Handle" relationType="IsPartOf">hdl:11858/00-1734-0000-0003-7666-B</datacite:relatedIdentifier>
</datacite:relatedIdentifiers>
<datacite:sizes>
<datacite:size>527 Bytes</datacite:size>
</datacite:sizes>
<datacite:formats>
<datacite:format>text/tg.edition+tg.aggregation+xml</datacite:format>
</datacite:formats>
<datacite:version>0</datacite:version>
<datacite:rightsList>
<datacite:rights rightsURI="http://creativecommons.org/licenses/by/3.0/de/legalcode"> Der annotierte Datenbestand der Digitalen Bibliothek inklusive
Metadaten sowie davon einzeln zugängliche Teile sind eine Abwandlung
des Datenbestandes von www.editura.de durch TextGrid und werden
unter der Lizenz Creative Commons Namensnennung 3.0 Deutschland
Lizenz (by-Nennung TextGrid) veröffentlicht. Die Lizenz bezieht sich
nicht auf die der Annotation zu Grunde liegenden allgemeinfreien
Texte (Siehe auch Punkt 2 der Lizenzbestimmungen).</datacite:rights>
<datacite:rights rightsURI="info:eu-repo/semantics/openAccess"/>
</datacite:rightsList>
<datacite:descriptions>
<datacite:description descriptionType="Abstract"/>
</datacite:descriptions>
<datacite:geoLocations>
<datacite:geoLocation>
<datacite:geoLocationPlace
xmlns:xs="http://www.w3.org/2001/XMLSchema" xsi:type="xs:string">Hamburg</datacite:geoLocationPlace>
</datacite:geoLocation>
</datacite:geoLocations>
</datacite:resource>
<oaf:identifier identifierType="handle">hdl:11858/00-1734-0000-0003-7664-F</oaf:identifier>
<dr:CobjCategory type="dataset">0021</dr:CobjCategory>
<oaf:refereed>0002</oaf:refereed>
<oaf:dateAccepted>2012-01-01</oaf:dateAccepted>
<oaf:accessrights>OPEN</oaf:accessrights>
<oaf:license>http://creativecommons.org/licenses/by/3.0/de/legalcode</oaf:license>
<oaf:language>und</oaf:language>
<oaf:hostedBy id="re3data_____::r3d100011365" name="TextGrid Repository"/>
<oaf:collectedFrom id="re3data_____::r3d100011365" name="TextGrid Repository"/>
</metadata>
<about xmlns:dc="http://purl.org/dc/elements/1.1/"
xmlns:dri="http://www.driver-repository.eu/namespace/dri" xmlns:prov="http://www.openarchives.org/OAI/2.0/provenance">
<provenance xmlns="http://www.openarchives.org/OAI/2.0/provenance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/provenance http://www.openarchives.org/OAI/2.0/provenance.xsd">
<originDescription altered="true" harvestDate="2020-11-17T09:34:11.128+01:00">
<baseURL>https%3A%2F%2Fdev.textgridlab.org%2F1.0%2Ftgoaipmh%2Foai</baseURL>
<identifier>textgrid:q9cv.0</identifier>
<datestamp>2012-01-21T13:35:20Z</datestamp>
<metadataNamespace>http://schema.datacite.org/oai/oai-1.0/</metadataNamespace>
</originDescription>
</provenance>
<oaf:datainfo>
<oaf:inferred>false</oaf:inferred>
<oaf:deletedbyinference>false</oaf:deletedbyinference>
<oaf:trust>0.9</oaf:trust>
<oaf:inferenceprovenance/>
<oaf:provenanceaction classid="sysimport:crosswalk"
classname="sysimport:crosswalk"
schemeid="dnet:provenanceActions" schemename="dnet:provenanceActions"/>
</oaf:datainfo>
</about>
</record>
Write Preview
Loading…
Cancel
Save