Examples with MonolingualTextValue org.wikidata.wdtk.datamodel.interfaces.MonolingualTextValue used on opensource projects

Example 16 with MonolingualTextValue

use of org.wikidata.wdtk.datamodel.interfaces.MonolingualTextValue in project OpenRefine by OpenRefine.

the class TermedStatementEntityEditTest method testMergeLabelsIfNewOverriding.

@Test
public void testMergeLabelsIfNewOverriding() {
    MonolingualTextValue label1 = Datamodel.makeMonolingualTextValue("first label", "en");
    MonolingualTextValue label2 = Datamodel.makeMonolingualTextValue("second label", "en");
    TermedStatementEntityEdit edit1 = new TermedStatementEntityEditBuilder(existingSubject).addLabel(label1, true).build();
    TermedStatementEntityEdit edit2 = new TermedStatementEntityEditBuilder(existingSubject).addLabel(label2, false).build();
    TermedStatementEntityEdit merged = edit1.merge(edit2);
    assertEquals(Collections.singleton(label1), merged.getLabels());
    assertEquals(Collections.emptySet(), merged.getLabelsIfNew());
}

Example 17 with MonolingualTextValue

use of org.wikidata.wdtk.datamodel.interfaces.MonolingualTextValue in project OpenRefine by OpenRefine.

the class TermedStatementEntityEditTest method testNormalizeTerms.

@Test
public void testNormalizeTerms() {
    MonolingualTextValue aliasEn = Datamodel.makeMonolingualTextValue("alias", "en");
    MonolingualTextValue aliasFr = Datamodel.makeMonolingualTextValue("coucou", "fr");
    TermedStatementEntityEdit updateA = new TermedStatementEntityEditBuilder(newSubject).addLabel(label, true).addAlias(aliasEn).addAlias(aliasFr).build();
    assertFalse(updateA.isNull());
    TermedStatementEntityEdit normalized = updateA.normalizeLabelsAndAliases();
    TermedStatementEntityEdit expectedUpdate = new TermedStatementEntityEditBuilder(newSubject).addLabel(label, true).addAlias(aliasEn).addLabel(aliasFr, true).build();
    assertEquals(expectedUpdate, normalized);
}

Example 18 with MonolingualTextValue

use of org.wikidata.wdtk.datamodel.interfaces.MonolingualTextValue in project OpenRefine by OpenRefine.

the class LaxValueMatcherTests method testMonolingualText.

@Test
public void testMonolingualText() {
    MonolingualTextValue value1 = Datamodel.makeMonolingualTextValue("foo", "en");
    MonolingualTextValue value2 = Datamodel.makeMonolingualTextValue("\tfoo ", "en");
    MonolingualTextValue value3 = Datamodel.makeMonolingualTextValue("bar", "en");
    assertTrue(SUT.match(value1, value2));
    assertFalse(SUT.match(value1, value3));
}

Example 19 with MonolingualTextValue

use of org.wikidata.wdtk.datamodel.interfaces.MonolingualTextValue in project OpenRefine by OpenRefine.

the class EditBatchProcessorTest method testMultipleBatchesMediaInfo.

@Test
public void testMultipleBatchesMediaInfo() throws MediaWikiApiErrorException, InterruptedException, IOException {
    // Prepare test data
    MonolingualTextValue label = Datamodel.makeMonolingualTextValue("village in Nepal", "en");
    List<MonolingualTextValue> labels = Collections.singletonList(label);
    TermUpdate labelsUpdate = Datamodel.makeTermUpdate(labels, Collections.emptyList());
    List<String> ids = new ArrayList<>();
    for (int i = 124; i < 190; i++) {
        ids.add("M" + String.valueOf(i));
    }
    List<MediaInfoIdValue> mids = ids.stream().map(e -> Datamodel.makeWikimediaCommonsMediaInfoIdValue(e)).collect(Collectors.toList());
    List<TermedStatementEntityEdit> batch = mids.stream().map(mid -> new TermedStatementEntityEditBuilder(mid).addLabel(label, false).build()).collect(Collectors.toList());
    int batchSize = 50;
    List<MediaInfoDocument> fullBatch = mids.stream().map(mid -> Datamodel.makeMediaInfoDocument(mid)).collect(Collectors.toList());
    List<MediaInfoDocument> firstBatch = fullBatch.subList(0, batchSize);
    List<MediaInfoDocument> secondBatch = fullBatch.subList(batchSize, fullBatch.size());
    when(fetcher.getEntityDocuments(toMids(firstBatch))).thenReturn(toMapMediaInfo(firstBatch));
    when(fetcher.getEntityDocuments(toMids(secondBatch))).thenReturn(toMapMediaInfo(secondBatch));
    // Run edits
    EditBatchProcessor processor = new EditBatchProcessor(fetcher, editor, batch, library, summary, maxlag, tags, batchSize);
    assertEquals(0, processor.progress());
    for (int i = 124; i < 190; i++) {
        assertEquals(processor.remainingEdits(), 190 - i);
        processor.performEdit();
    }
    assertEquals(0, processor.remainingEdits());
    assertEquals(100, processor.progress());
    // Check result
    assertEquals(new NewEntityLibrary(), library);
    verify(fetcher, times(1)).getEntityDocuments(toMids(firstBatch));
    verify(fetcher, times(1)).getEntityDocuments(toMids(secondBatch));
    for (MediaInfoDocument doc : fullBatch) {
        StatementUpdate statementUpdate = Datamodel.makeStatementUpdate(Collections.emptyList(), Collections.emptyList(), Collections.emptyList());
        verify(editor, times(1)).editEntityDocument(Datamodel.makeMediaInfoUpdate((MediaInfoIdValue) doc.getEntityId(), doc.getRevisionId(), labelsUpdate, statementUpdate), false, summary, tags);
    }
}

Example 20 with MonolingualTextValue

use of org.wikidata.wdtk.datamodel.interfaces.MonolingualTextValue in project OpenRefine by OpenRefine.

the class LaxValueMatcher method match.

@Override
public boolean match(Value existing, Value added) {
    if (existing instanceof EntityIdValue && added instanceof EntityIdValue) {
        // to mix up entities from different Wikibases in the same data slot
        return ((EntityIdValue) existing).getId().equals(((EntityIdValue) added).getId());
    } else if (existing instanceof StringValue && added instanceof StringValue) {
        // disregard trailing whitespace differences
        String existingStr = ((StringValue) existing).getString().trim();
        String addedStr = ((StringValue) added).getString().trim();
        // if they look like URLs, then http(s) and trailing slashes do not matter
        try {
            URI existingUrl = extraURINormalize(new URI(existingStr).normalize());
            URI addedUrl = extraURINormalize(new URI(addedStr).normalize());
            return existingUrl.equals(addedUrl);
        } catch (URISyntaxException e) {
            // fall back on basic comparison
            ;
        }
        return existingStr.equals(addedStr);
    } else if (existing instanceof MonolingualTextValue && added instanceof MonolingualTextValue) {
        // ignore differences of trailing whitespace
        MonolingualTextValue existingMTV = (MonolingualTextValue) existing;
        MonolingualTextValue addedMTV = (MonolingualTextValue) added;
        return (existingMTV.getLanguageCode().equals(addedMTV.getLanguageCode()) && existingMTV.getText().trim().equals(addedMTV.getText().trim()));
    } else if (existing instanceof QuantityValue && added instanceof QuantityValue) {
        QuantityValue existingQuantity = (QuantityValue) existing;
        QuantityValue addedQuantity = (QuantityValue) added;
        BigDecimal existingLowerBound = existingQuantity.getLowerBound();
        BigDecimal addedLowerBound = addedQuantity.getLowerBound();
        BigDecimal existingUpperBound = existingQuantity.getUpperBound();
        BigDecimal addedUpperBound = addedQuantity.getUpperBound();
        // artificially set bounds for quantities which have neither lower nor upper bounds
        if (existingLowerBound == null && existingUpperBound == null) {
            existingLowerBound = existingQuantity.getNumericValue();
            existingUpperBound = existingQuantity.getNumericValue();
        }
        if (addedLowerBound == null && addedUpperBound == null) {
            addedLowerBound = addedQuantity.getNumericValue();
            addedUpperBound = addedQuantity.getNumericValue();
        }
        if (existingQuantity.getUnit().equals(addedQuantity.getUnit()) && (existingLowerBound != null) && (addedLowerBound != null) && (existingUpperBound != null) && (addedUpperBound != null)) {
            // Consider the two values to be equal when their confidence interval overlaps
            return ((existingLowerBound.compareTo(addedLowerBound) <= 0 && addedLowerBound.compareTo(existingUpperBound) <= 0) || (addedLowerBound.compareTo(existingLowerBound) <= 0 && existingLowerBound.compareTo(addedUpperBound) <= 0));
        }
    } else if (existing instanceof GlobeCoordinatesValue && added instanceof GlobeCoordinatesValue) {
        GlobeCoordinatesValue addedCoords = (GlobeCoordinatesValue) added;
        GlobeCoordinatesValue existingCoords = (GlobeCoordinatesValue) existing;
        if (!addedCoords.getGlobeItemId().getId().equals(existingCoords.getGlobeItemId().getId())) {
            return false;
        }
        double addedMinLon = addedCoords.getLongitude() - addedCoords.getPrecision();
        double addedMaxLon = addedCoords.getLongitude() + addedCoords.getPrecision();
        double addedMinLat = addedCoords.getLatitude() - addedCoords.getPrecision();
        double addedMaxLat = addedCoords.getLatitude() + addedCoords.getPrecision();
        double existingMinLon = existingCoords.getLongitude() - existingCoords.getPrecision();
        double existingMaxLon = existingCoords.getLongitude() + existingCoords.getPrecision();
        double existingMinLat = existingCoords.getLatitude() - existingCoords.getPrecision();
        double existingMaxLat = existingCoords.getLatitude() + existingCoords.getPrecision();
        // return true when the two "rectangles" (in coordinate space) overlap (not strictly)
        return ((addedMinLon <= existingMinLon && addedMinLat <= existingMinLat && existingMinLon <= addedMaxLon && existingMinLat <= addedMaxLat) || (existingMinLon <= addedMinLon && existingMinLat <= addedMinLat && addedMinLon <= existingMaxLon && addedMinLat <= existingMaxLat));
    } else if (existing instanceof TimeValue && added instanceof TimeValue) {
        TimeValue existingTime = (TimeValue) existing;
        TimeValue addedTime = (TimeValue) added;
        if (!existingTime.getPreferredCalendarModel().equals(addedTime.getPreferredCalendarModel())) {
            return false;
        }
        int minPrecision = Math.min(existingTime.getPrecision(), addedTime.getPrecision());
        if (minPrecision <= 9) {
            // the precision is a multiple of years
            long yearPrecision = (long) Math.pow(10, 9 - minPrecision);
            long addedValue = addedTime.getYear() / yearPrecision;
            long existingValue = existingTime.getYear() / yearPrecision;
            return addedValue == existingValue;
        } else if (minPrecision == 10) {
            // month precision
            return (addedTime.getYear() == existingTime.getYear() && addedTime.getMonth() == existingTime.getMonth());
        } else if (minPrecision == 11) {
            // day precision
            return (addedTime.getYear() == existingTime.getYear() && addedTime.getMonth() == existingTime.getMonth() && addedTime.getDay() == existingTime.getDay());
        }
    // TODO possible improvements: bounds support, timezone support
    }
    // fall back to exact comparison for other datatypes
    return existing.equals(added);
}