Examples with SolrInputField org.apache.solr.common.SolrInputField used on opensource projects

Example 1 with SolrInputField

use of org.apache.solr.common.SolrInputField in project lucene-solr by apache.

the class LanguageIdentifierUpdateProcessor method process.

/**
   * This is the main, testable process method called from processAdd()
   * @param doc the SolrInputDocument to work on
   * @return the modified SolrInputDocument
   */
protected SolrInputDocument process(SolrInputDocument doc) {
    String docLang = null;
    HashSet<String> docLangs = new HashSet<>();
    String fallbackLang = getFallbackLang(doc, fallbackFields, fallbackValue);
    if (langField == null || !doc.containsKey(langField) || (doc.containsKey(langField) && overwrite)) {
        List<DetectedLanguage> languagelist = detectLanguage(doc);
        docLang = resolveLanguage(languagelist, fallbackLang);
        docLangs.add(docLang);
        log.debug("Detected main document language from fields " + Arrays.toString(inputFields) + ": " + docLang);
        if (doc.containsKey(langField) && overwrite) {
            log.debug("Overwritten old value " + doc.getFieldValue(langField));
        }
        if (langField != null && langField.length() != 0) {
            doc.setField(langField, docLang);
        }
    } else {
        // langField is set, we sanity check it against whitelist and fallback
        docLang = resolveLanguage((String) doc.getFieldValue(langField), fallbackLang);
        docLangs.add(docLang);
        log.debug("Field " + langField + " already contained value " + docLang + ", not overwriting.");
    }
    if (enableMapping) {
        for (String fieldName : allMapFieldsSet) {
            if (doc.containsKey(fieldName)) {
                String fieldLang;
                if (mapIndividual && mapIndividualFieldsSet.contains(fieldName)) {
                    List<DetectedLanguage> languagelist = detectLanguage(doc);
                    fieldLang = resolveLanguage(languagelist, docLang);
                    docLangs.add(fieldLang);
                    log.debug("Mapping field " + fieldName + " using individually detected language " + fieldLang);
                } else {
                    fieldLang = docLang;
                    log.debug("Mapping field " + fieldName + " using document global language " + fieldLang);
                }
                String mappedOutputField = getMappedField(fieldName, fieldLang);
                if (mappedOutputField != null) {
                    log.debug("Mapping field {} to {}", doc.getFieldValue(docIdField), fieldLang);
                    SolrInputField inField = doc.getField(fieldName);
                    doc.setField(mappedOutputField, inField.getValue());
                    if (!mapKeepOrig) {
                        log.debug("Removing old field {}", fieldName);
                        doc.removeField(fieldName);
                    }
                } else {
                    throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, "Invalid output field mapping for " + fieldName + " field and language: " + fieldLang);
                }
            }
        }
    }
    // Set the languages field to an array of all detected languages
    if (langsField != null && langsField.length() != 0) {
        doc.setField(langsField, docLangs.toArray());
    }
    return doc;
}

Example 2 with SolrInputField

use of org.apache.solr.common.SolrInputField in project lucene-solr by apache.

the class DocumentBuilder method toDocument.

/**
   * Convert a SolrInputDocument to a lucene Document.
   * 
   * This function should go elsewhere.  This builds the Document without an
   * extra Map<> checking for multiple values.  For more discussion, see:
   * http://www.nabble.com/Re%3A-svn-commit%3A-r547493---in--lucene-solr-trunk%3A-.--src-java-org-apache-solr-common--src-java-org-apache-solr-schema--src-java-org-apache-solr-update--src-test-org-apache-solr-common--tf3931539.html
   * 
   * TODO: /!\ NOTE /!\ This semantics of this function are still in flux.  
   * Something somewhere needs to be able to fill up a SolrDocument from
   * a lucene document - this is one place that may happen.  It may also be
   * moved to an independent function
   * 
   * @since solr 1.3
   * 
   * @param doc SolrInputDocument from which the document has to be built
   * @param schema Schema instance
   * @param forInPlaceUpdate Whether the output document would be used for an in-place update or not. When this is true,
   *        default fields values and copy fields targets are not populated.
   * @return Built Lucene document

   */
public static Document toDocument(SolrInputDocument doc, IndexSchema schema, boolean forInPlaceUpdate) {
    final SchemaField uniqueKeyField = schema.getUniqueKeyField();
    final String uniqueKeyFieldName = null == uniqueKeyField ? null : uniqueKeyField.getName();
    Document out = new Document();
    Set<String> usedFields = Sets.newHashSet();
    // Load fields from SolrDocument to Document
    for (SolrInputField field : doc) {
        String name = field.getName();
        SchemaField sfield = schema.getFieldOrNull(name);
        boolean used = false;
        // Make sure it has the correct number
        if (sfield != null && !sfield.multiValued() && field.getValueCount() > 1) {
            throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, "ERROR: " + getID(doc, schema) + "multiple values encountered for non multiValued field " + sfield.getName() + ": " + field.getValue());
        }
        List<CopyField> copyFields = schema.getCopyFieldsList(name);
        if (copyFields.size() == 0)
            copyFields = null;
        // load each field value
        boolean hasField = false;
        try {
            for (Object v : field) {
                if (v == null) {
                    continue;
                }
                hasField = true;
                if (sfield != null) {
                    used = true;
                    addField(out, sfield, v, name.equals(uniqueKeyFieldName) ? false : forInPlaceUpdate);
                    // record the field as having a value
                    usedFields.add(sfield.getName());
                }
                // This could happen whether it is explicit or not.
                if (copyFields != null) {
                    // and this is the uniqueKey field (because the uniqueKey can't change so no need to "update" the copyField).
                    if (!(forInPlaceUpdate && name.equals(uniqueKeyFieldName))) {
                        for (CopyField cf : copyFields) {
                            SchemaField destinationField = cf.getDestination();
                            final boolean destHasValues = usedFields.contains(destinationField.getName());
                            // check if the copy field is a multivalued or not
                            if (!destinationField.multiValued() && destHasValues) {
                                throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, "ERROR: " + getID(doc, schema) + "multiple values encountered for non multiValued copy field " + destinationField.getName() + ": " + v);
                            }
                            used = true;
                            // Perhaps trim the length of a copy field
                            Object val = v;
                            if (val instanceof String && cf.getMaxChars() > 0) {
                                val = cf.getLimitedValue((String) val);
                            }
                            addField(out, destinationField, val, destinationField.getName().equals(uniqueKeyFieldName) ? false : forInPlaceUpdate);
                            // record the field as having a value
                            usedFields.add(destinationField.getName());
                        }
                    }
                }
            }
        } catch (SolrException ex) {
            throw ex;
        } catch (Exception ex) {
            throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, "ERROR: " + getID(doc, schema) + "Error adding field '" + field.getName() + "'='" + field.getValue() + "' msg=" + ex.getMessage(), ex);
        }
        // make sure the field was used somehow...
        if (!used && hasField) {
            throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, "ERROR: " + getID(doc, schema) + "unknown field '" + name + "'");
        }
    }
    // during the full indexing initially.
    if (!forInPlaceUpdate) {
        for (SchemaField field : schema.getRequiredFields()) {
            if (out.getField(field.getName()) == null) {
                if (field.getDefaultValue() != null) {
                    addField(out, field, field.getDefaultValue(), false);
                } else {
                    String msg = getID(doc, schema) + "missing required field: " + field.getName();
                    throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, msg);
                }
            }
        }
    }
    if (!forInPlaceUpdate) {
        moveLargestFieldLast(out);
    }
    return out;
}

Example 3 with SolrInputField

use of org.apache.solr.common.SolrInputField in project lucene-solr by apache.

the class DocumentBuilderTest method testSolrInputFieldEquality.

public void testSolrInputFieldEquality() {
    String randomString = TestUtil.randomSimpleString(random(), 10, 20);
    int val = random().nextInt();
    SolrInputField sif1 = new SolrInputField(randomString);
    sif1.setValue(val);
    SolrInputField sif2 = new SolrInputField(randomString);
    sif2.setValue(val);
    assertTrue(assertSolrInputFieldEquals(sif1, sif2));
    sif2.setName("foo");
    assertFalse(assertSolrInputFieldEquals(sif1, sif2));
}

Example 4 with SolrInputField

use of org.apache.solr.common.SolrInputField in project lucene-solr by apache.

the class TestDocumentObjectBinder method testSimple.

public void testSimple() throws Exception {
    DocumentObjectBinder binder = new DocumentObjectBinder();
    XMLResponseParser parser = new XMLResponseParser();
    NamedList<Object> nl = parser.processResponse(new StringReader(xml));
    QueryResponse res = new QueryResponse(nl, null);
    SolrDocumentList solDocList = res.getResults();
    List<Item> l = binder.getBeans(Item.class, res.getResults());
    assertEquals(solDocList.size(), l.size());
    assertEquals(solDocList.get(0).getFieldValue("features"), l.get(0).features);
    Item item = new Item();
    item.id = "aaa";
    item.categories = new String[] { "aaa", "bbb", "ccc" };
    SolrInputDocument out = binder.toSolrInputDocument(item);
    assertEquals(item.id, out.getFieldValue("id"));
    SolrInputField catfield = out.getField("cat");
    assertEquals(3, catfield.getValueCount());
    List<String> catValues = (List<String>) catfield.getValue();
    assertEquals("aaa", catValues.get(0));
    assertEquals("bbb", catValues.get(1));
    assertEquals("ccc", catValues.get(2));
}

Example 5 with SolrInputField

use of org.apache.solr.common.SolrInputField in project lucene-solr by apache.

the class TestUpdateRequestCodec method testIteratable.

@Test
public void testIteratable() throws IOException {
    final List<String> values = new ArrayList<>();
    values.add("iterItem1");
    values.add("iterItem2");
    UpdateRequest updateRequest = new UpdateRequest();
    updateRequest.deleteByQuery("*:*");
    SolrInputDocument doc = new SolrInputDocument();
    doc.addField("id", 1);
    doc.addField("desc", "one");
    // imagine someone adding a custom Bean that implements Iterable 
    // but is not a Collection
    doc.addField("iter", new Iterable<String>() {

        @Override
        public Iterator<String> iterator() {
            return values.iterator();
        }
    });
    doc.addField("desc", "1");
    updateRequest.add(doc);
    JavaBinUpdateRequestCodec codec = new JavaBinUpdateRequestCodec();
    ByteArrayOutputStream baos = new ByteArrayOutputStream();
    codec.marshal(updateRequest, baos);
    final List<SolrInputDocument> docs = new ArrayList<>();
    JavaBinUpdateRequestCodec.StreamingUpdateHandler handler = new JavaBinUpdateRequestCodec.StreamingUpdateHandler() {

        @Override
        public void update(SolrInputDocument document, UpdateRequest req, Integer commitWithin, Boolean overwrite) {
            Assert.assertNotNull(req.getParams());
            docs.add(document);
        }
    };
    UpdateRequest updateUnmarshalled = codec.unmarshal(new ByteArrayInputStream(baos.toByteArray()), handler);
    for (SolrInputDocument document : docs) {
        updateUnmarshalled.add(document);
    }
    SolrInputDocument outDoc = updateUnmarshalled.getDocuments().get(0);
    SolrInputField iter = outDoc.getField("iter");
    Assert.assertNotNull("iter field is null", iter);
    Object iterVal = iter.getValue();
    Assert.assertTrue("iterVal is not a Collection", iterVal instanceof Collection);
    Assert.assertEquals("iterVal contents", values, iterVal);
}