Examples with SolrCore org.apache.solr.core.SolrCore used on opensource projects

Example 1 with SolrCore

use of org.apache.solr.core.SolrCore in project lucene-solr by apache.

the class XPathEntityProcessor method initXpathReader.

private void initXpathReader(VariableResolver resolver) {
    reinitXPathReader = false;
    useSolrAddXml = Boolean.parseBoolean(context.getEntityAttribute(USE_SOLR_ADD_SCHEMA));
    streamRows = Boolean.parseBoolean(context.getEntityAttribute(STREAM));
    if (context.getResolvedEntityAttribute("batchSize") != null) {
        blockingQueueSize = Integer.parseInt(context.getEntityAttribute("batchSize"));
    }
    if (context.getResolvedEntityAttribute("readTimeOut") != null) {
        blockingQueueTimeOut = Integer.parseInt(context.getEntityAttribute("readTimeOut"));
    }
    String xslt = context.getEntityAttribute(XSL);
    if (xslt != null) {
        xslt = context.replaceTokens(xslt);
        try {
            // create an instance of TransformerFactory
            TransformerFactory transFact = TransformerFactory.newInstance();
            final SolrCore core = context.getSolrCore();
            final StreamSource xsltSource;
            if (core != null) {
                final ResourceLoader loader = core.getResourceLoader();
                transFact.setURIResolver(new SystemIdResolver(loader).asURIResolver());
                xsltSource = new StreamSource(loader.openResource(xslt), SystemIdResolver.createSystemIdFromResourceName(xslt));
            } else {
                // fallback for tests
                xsltSource = new StreamSource(xslt);
            }
            transFact.setErrorListener(xmllog);
            try {
                xslTransformer = transFact.newTransformer(xsltSource);
            } finally {
                // some XML parsers are broken and don't close the byte stream (but they should according to spec)
                IOUtils.closeQuietly(xsltSource.getInputStream());
            }
            LOG.info("Using xslTransformer: " + xslTransformer.getClass().getName());
        } catch (Exception e) {
            throw new DataImportHandlerException(SEVERE, "Error initializing XSL ", e);
        }
    }
    if (useSolrAddXml) {
        // Support solr add documents
        xpathReader = new XPathRecordReader("/add/doc");
        xpathReader.addField("name", "/add/doc/field/@name", true);
        xpathReader.addField("value", "/add/doc/field", true);
    } else {
        String forEachXpath = context.getResolvedEntityAttribute(FOR_EACH);
        if (forEachXpath == null)
            throw new DataImportHandlerException(SEVERE, "Entity : " + context.getEntityAttribute("name") + " must have a 'forEach' attribute");
        if (forEachXpath.equals(context.getEntityAttribute(FOR_EACH)))
            reinitXPathReader = true;
        try {
            xpathReader = new XPathRecordReader(forEachXpath);
            for (Map<String, String> field : context.getAllEntityFields()) {
                if (field.get(XPATH) == null)
                    continue;
                int flags = 0;
                if ("true".equals(field.get("flatten"))) {
                    flags = XPathRecordReader.FLATTEN;
                }
                String xpath = field.get(XPATH);
                xpath = context.replaceTokens(xpath);
                //for each xml
                if (!xpath.equals(field.get(XPATH)) && !context.isRootEntity())
                    reinitXPathReader = true;
                xpathReader.addField(field.get(DataImporter.COLUMN), xpath, Boolean.parseBoolean(field.get(DataImporter.MULTI_VALUED)), flags);
            }
        } catch (RuntimeException e) {
            throw new DataImportHandlerException(SEVERE, "Exception while reading xpaths for fields", e);
        }
    }
    String url = context.getEntityAttribute(URL);
    List<String> l = url == null ? Collections.EMPTY_LIST : resolver.getVariables(url);
    for (String s : l) {
        if (s.startsWith(entityName + ".")) {
            if (placeHolderVariables == null)
                placeHolderVariables = new ArrayList<>();
            placeHolderVariables.add(s.substring(entityName.length() + 1));
        }
    }
    for (Map<String, String> fld : context.getAllEntityFields()) {
        if (fld.get(COMMON_FIELD) != null && "true".equals(fld.get(COMMON_FIELD))) {
            if (commonFields == null)
                commonFields = new ArrayList<>();
            commonFields.add(fld.get(DataImporter.COLUMN));
        }
    }
}

Example 2 with SolrCore

use of org.apache.solr.core.SolrCore in project lucene-solr by apache.

the class UIMAUpdateRequestProcessorTest method testProcessorConfiguration.

@Test
public void testProcessorConfiguration() {
    SolrCore core = h.getCore();
    UpdateRequestProcessorChain chained = core.getUpdateProcessingChain(UIMA_CHAIN);
    assertNotNull(chained);
    UIMAUpdateRequestProcessorFactory factory = (UIMAUpdateRequestProcessorFactory) chained.getProcessors().get(0);
    assertNotNull(factory);
    UpdateRequestProcessor processor = factory.getInstance(req(), null, null);
    assertTrue(processor instanceof UIMAUpdateRequestProcessor);
}

Example 3 with SolrCore

use of org.apache.solr.core.SolrCore in project lucene-solr by apache.

the class SearchHandler method inform.

/**
   * Initialize the components based on name.  Note, if using <code>INIT_FIRST_COMPONENTS</code> or <code>INIT_LAST_COMPONENTS</code>,
   * then the {@link DebugComponent} will always occur last.  If this is not desired, then one must explicitly declare all components using
   * the <code>INIT_COMPONENTS</code> syntax.
   */
@Override
@SuppressWarnings("unchecked")
public void inform(SolrCore core) {
    this.core = core;
    Set<String> missing = new HashSet<>();
    List<String> c = (List<String>) initArgs.get(INIT_COMPONENTS);
    missing.addAll(core.getSearchComponents().checkContains(c));
    List<String> first = (List<String>) initArgs.get(INIT_FIRST_COMPONENTS);
    missing.addAll(core.getSearchComponents().checkContains(first));
    List<String> last = (List<String>) initArgs.get(INIT_LAST_COMPONENTS);
    missing.addAll(core.getSearchComponents().checkContains(last));
    if (!missing.isEmpty())
        throw new SolrException(SolrException.ErrorCode.SERVER_ERROR, "Missing SearchComponents named : " + missing);
    if (c != null && (first != null || last != null))
        throw new SolrException(SolrException.ErrorCode.SERVER_ERROR, "First/Last components only valid if you do not declare 'components'");
    if (shfInfo == null) {
        shardHandlerFactory = core.getCoreContainer().getShardHandlerFactory();
    } else {
        shardHandlerFactory = core.createInitInstance(shfInfo, ShardHandlerFactory.class, null, null);
        core.addCloseHook(new CloseHook() {

            @Override
            public void preClose(SolrCore core) {
                shardHandlerFactory.close();
            }

            @Override
            public void postClose(SolrCore core) {
            }
        });
    }
}

Example 4 with SolrCore

use of org.apache.solr.core.SolrCore in project lucene-solr by apache.

the class CarrotClusteringEngine method getDocuments.

/**
   * Prepares Carrot2 documents for clustering.
   */
private List<Document> getDocuments(SolrDocumentList solrDocList, Map<SolrDocument, Integer> docIds, Query query, final SolrQueryRequest sreq) throws IOException {
    SolrHighlighter highlighter = null;
    SolrParams solrParams = sreq.getParams();
    SolrCore core = sreq.getCore();
    String urlField = solrParams.get(CarrotParams.URL_FIELD_NAME, "url");
    String titleFieldSpec = solrParams.get(CarrotParams.TITLE_FIELD_NAME, "title");
    String snippetFieldSpec = solrParams.get(CarrotParams.SNIPPET_FIELD_NAME, titleFieldSpec);
    String languageField = solrParams.get(CarrotParams.LANGUAGE_FIELD_NAME, null);
    // Maps Solr field names to Carrot2 custom field names
    Map<String, String> customFields = getCustomFieldsMap(solrParams);
    // Parse language code map string into a map
    Map<String, String> languageCodeMap = new HashMap<>();
    if (StringUtils.isNotBlank(languageField)) {
        for (String pair : solrParams.get(CarrotParams.LANGUAGE_CODE_MAP, "").split("[, ]")) {
            final String[] split = pair.split(":");
            if (split.length == 2 && StringUtils.isNotBlank(split[0]) && StringUtils.isNotBlank(split[1])) {
                languageCodeMap.put(split[0], split[1]);
            } else {
                log.warn("Unsupported format for " + CarrotParams.LANGUAGE_CODE_MAP + ": '" + pair + "'. Skipping this mapping.");
            }
        }
    }
    // Get the documents
    boolean produceSummary = solrParams.getBool(CarrotParams.PRODUCE_SUMMARY, false);
    SolrQueryRequest req = null;
    String[] snippetFieldAry = null;
    if (produceSummary) {
        highlighter = HighlightComponent.getHighlighter(core);
        if (highlighter != null) {
            Map<String, Object> args = new HashMap<>();
            snippetFieldAry = snippetFieldSpec.split("[, ]");
            args.put(HighlightParams.FIELDS, snippetFieldAry);
            args.put(HighlightParams.HIGHLIGHT, "true");
            //we don't care about actually highlighting the area
            args.put(HighlightParams.SIMPLE_PRE, "");
            args.put(HighlightParams.SIMPLE_POST, "");
            args.put(HighlightParams.FRAGSIZE, solrParams.getInt(CarrotParams.SUMMARY_FRAGSIZE, solrParams.getInt(HighlightParams.FRAGSIZE, 100)));
            args.put(HighlightParams.SNIPPETS, solrParams.getInt(CarrotParams.SUMMARY_SNIPPETS, solrParams.getInt(HighlightParams.SNIPPETS, 1)));
            req = new LocalSolrQueryRequest(core, query.toString(), "", 0, 1, args) {

                @Override
                public SolrIndexSearcher getSearcher() {
                    return sreq.getSearcher();
                }
            };
        } else {
            log.warn("No highlighter configured, cannot produce summary");
            produceSummary = false;
        }
    }
    Iterator<SolrDocument> docsIter = solrDocList.iterator();
    List<Document> result = new ArrayList<>(solrDocList.size());
    float[] scores = { 1.0f };
    int[] docsHolder = new int[1];
    Query theQuery = query;
    while (docsIter.hasNext()) {
        SolrDocument sdoc = docsIter.next();
        String snippet = null;
        // See comment in ClusteringComponent#finishStage().
        if (produceSummary && docIds != null) {
            docsHolder[0] = docIds.get(sdoc).intValue();
            DocList docAsList = new DocSlice(0, 1, docsHolder, scores, 1, 1.0f);
            NamedList<Object> highlights = highlighter.doHighlighting(docAsList, theQuery, req, snippetFieldAry);
            if (highlights != null && highlights.size() == 1) {
                // should only be one value given our setup
                // should only be one document
                @SuppressWarnings("unchecked") NamedList<String[]> tmp = (NamedList<String[]>) highlights.getVal(0);
                final StringBuilder sb = new StringBuilder();
                for (int j = 0; j < snippetFieldAry.length; j++) {
                    // Join fragments with a period, so that Carrot2 does not create
                    // cross-fragment phrases, such phrases rarely make sense.
                    String[] highlt = tmp.get(snippetFieldAry[j]);
                    if (highlt != null && highlt.length > 0) {
                        for (int i = 0; i < highlt.length; i++) {
                            sb.append(highlt[i]);
                            sb.append(" . ");
                        }
                    }
                }
                snippet = sb.toString();
            }
        }
        // If summaries not enabled or summary generation failed, use full content.
        if (snippet == null) {
            snippet = getConcatenated(sdoc, snippetFieldSpec);
        }
        // Create a Carrot2 document
        Document carrotDocument = new Document(getConcatenated(sdoc, titleFieldSpec), snippet, ObjectUtils.toString(sdoc.getFieldValue(urlField), ""));
        // Store Solr id of the document, we need it to map document instances 
        // found in clusters back to identifiers.
        carrotDocument.setField(SOLR_DOCUMENT_ID, sdoc.getFieldValue(idFieldName));
        // Set language
        if (StringUtils.isNotBlank(languageField)) {
            Collection<Object> languages = sdoc.getFieldValues(languageField);
            if (languages != null) {
                // Use the first Carrot2-supported language
                for (Object l : languages) {
                    String lang = ObjectUtils.toString(l, "");
                    if (languageCodeMap.containsKey(lang)) {
                        lang = languageCodeMap.get(lang);
                    }
                    // language variants, such as 'zh-cn', but Carrot2 uses underscores.
                    if (lang.indexOf('-') > 0) {
                        lang = lang.replace('-', '_');
                    }
                    // If the language is supported by Carrot2, we'll get a non-null value
                    final LanguageCode carrot2Language = LanguageCode.forISOCode(lang);
                    if (carrot2Language != null) {
                        carrotDocument.setLanguage(carrot2Language);
                        break;
                    }
                }
            }
        }
        // Add custom fields
        if (customFields != null) {
            for (Entry<String, String> entry : customFields.entrySet()) {
                carrotDocument.setField(entry.getValue(), sdoc.getFieldValue(entry.getKey()));
            }
        }
        result.add(carrotDocument);
    }
    return result;
}

Example 5 with SolrCore

use of org.apache.solr.core.SolrCore in project lucene-solr by apache.

the class ClusteringComponentTest method testComponent.

@Test
public void testComponent() throws Exception {
    SolrCore core = h.getCore();
    SearchComponent sc = core.getSearchComponent("clustering");
    assertTrue("sc is null and it shouldn't be", sc != null);
    ModifiableSolrParams params = new ModifiableSolrParams();
    params.add(ClusteringComponent.COMPONENT_NAME, "true");
    params.add(CommonParams.Q, "*:*");
    params.add(ClusteringParams.USE_SEARCH_RESULTS, "true");
    SolrRequestHandler handler = core.getRequestHandler("standard");
    SolrQueryResponse rsp;
    rsp = new SolrQueryResponse();
    rsp.addResponseHeader(new SimpleOrderedMap<>());
    SolrQueryRequest req = new LocalSolrQueryRequest(core, params);
    handler.handleRequest(req, rsp);
    NamedList<?> values = rsp.getValues();
    Object clusters = values.get("clusters");
    //System.out.println("Clusters: " + clusters);
    assertTrue("clusters is null and it shouldn't be", clusters != null);
    req.close();
    params = new ModifiableSolrParams();
    params.add(ClusteringComponent.COMPONENT_NAME, "true");
    params.add(ClusteringParams.ENGINE_NAME, "mock");
    params.add(ClusteringParams.USE_COLLECTION, "true");
    params.add(QueryComponent.COMPONENT_NAME, "false");
    handler = core.getRequestHandler("docClustering");
    rsp = new SolrQueryResponse();
    rsp.addResponseHeader(new SimpleOrderedMap<>());
    req = new LocalSolrQueryRequest(core, params);
    handler.handleRequest(req, rsp);
    values = rsp.getValues();
    clusters = values.get("clusters");
    //System.out.println("Clusters: " + clusters);
    assertTrue("clusters is null and it shouldn't be", clusters != null);
    req.close();
}