Examples with SpanPayloadCheckQuery org.apache.lucene.queries.payloads.SpanPayloadCheckQuery used on opensource projects

Example 1 with SpanPayloadCheckQuery

use of org.apache.lucene.queries.payloads.SpanPayloadCheckQuery in project lucene-solr by apache.

the class PayloadCheckQParserPlugin method createParser.

@Override
public QParser createParser(String qstr, SolrParams localParams, SolrParams params, SolrQueryRequest req) {
    return new QParser(qstr, localParams, params, req) {

        @Override
        public Query parse() throws SyntaxError {
            String field = localParams.get(QueryParsing.F);
            String value = localParams.get(QueryParsing.V);
            String p = localParams.get("payloads");
            if (field == null) {
                throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, "'f' not specified");
            }
            if (value == null) {
                throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, "query string missing");
            }
            if (p == null) {
                throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, "'payloads' not specified");
            }
            FieldType ft = req.getCore().getLatestSchema().getFieldType(field);
            Analyzer analyzer = ft.getQueryAnalyzer();
            SpanQuery query = null;
            try {
                query = PayloadUtils.createSpanQuery(field, value, analyzer);
            } catch (IOException e) {
                throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, e);
            }
            if (query == null) {
                throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, "SpanQuery is null");
            }
            PayloadEncoder encoder = null;
            String e = PayloadUtils.getPayloadEncoder(ft);
            if ("float".equals(e)) {
                // TODO: centralize this string->PayloadEncoder logic (see DelimitedPayloadTokenFilterFactory)
                encoder = new FloatEncoder();
            } else if ("integer".equals(e)) {
                encoder = new IntegerEncoder();
            } else if ("identity".equals(e)) {
                encoder = new IdentityEncoder();
            }
            if (encoder == null) {
                throw new SolrException(SolrException.ErrorCode.BAD_REQUEST, "invalid encoder: " + e + " for field: " + field);
            }
            List<BytesRef> payloads = new ArrayList<>();
            // since payloads (most likely) came in whitespace delimited, just split
            String[] rawPayloads = p.split(" ");
            for (String rawPayload : rawPayloads) {
                if (rawPayload.length() > 0)
                    payloads.add(encoder.encode(rawPayload.toCharArray()));
            }
            return new SpanPayloadCheckQuery(query, payloads);
        }
    };
}

Example 2 with SpanPayloadCheckQuery

use of org.apache.lucene.queries.payloads.SpanPayloadCheckQuery in project lucene-solr by apache.

the class HighlighterTest method payloadFilteringSpanQuery.

@Test
public void payloadFilteringSpanQuery() throws IOException {
    clearIndex();
    String FIELD_NAME = "payloadDelimited";
    assertU(adoc("id", "0", FIELD_NAME, "word|7 word|2"));
    assertU(commit());
    //We search at a lower level than typical Solr tests because there's no QParser for payloads
    //Create query matching this payload
    Query query = new SpanPayloadCheckQuery(new SpanTermQuery(new Term(FIELD_NAME, "word")), //bytes for integer 7
    Collections.singletonList(new BytesRef(new byte[] { 0, 0, 0, 7 })));
    //invoke highlight component... the hard way
    final SearchComponent hlComp = h.getCore().getSearchComponent("highlight");
    SolrQueryRequest req = req("hl", "true", "hl.fl", FIELD_NAME, HighlightParams.USE_PHRASE_HIGHLIGHTER, "true");
    try {
        SolrQueryResponse resp = new SolrQueryResponse();
        ResponseBuilder rb = new ResponseBuilder(req, resp, Collections.singletonList(hlComp));
        rb.setHighlightQuery(query);
        rb.setResults(req.getSearcher().getDocListAndSet(query, (DocSet) null, null, 0, 1));
        //highlight:
        hlComp.prepare(rb);
        hlComp.process(rb);
        //inspect response
        final String[] snippets = (String[]) resp.getValues().findRecursive("highlighting", "0", FIELD_NAME);
        assertEquals("<em>word|7</em> word|2", snippets[0]);
    } finally {
        req.close();
    }
}

Example 3 with SpanPayloadCheckQuery

use of org.apache.lucene.queries.payloads.SpanPayloadCheckQuery in project lucene-solr by apache.

the class SynonymTokenizer method testPayloadQuery.

/** We can highlight based on payloads. It's supported both via term vectors and MemoryIndex since Lucene 5. */
public void testPayloadQuery() throws IOException, InvalidTokenOffsetsException {
    //"words" at positions 1 & 4
    final String text = "random words and words";
    //sets payload to "pos: X" (where X is position #)
    Analyzer analyzer = new MockPayloadAnalyzer();
    try (IndexWriter writer = new IndexWriter(dir, new IndexWriterConfig(analyzer))) {
        writer.deleteAll();
        Document doc = new Document();
        doc.add(new Field(FIELD_NAME, text, fieldType));
        writer.addDocument(doc);
        writer.commit();
    }
    try (IndexReader reader = DirectoryReader.open(dir)) {
        Query query = new SpanPayloadCheckQuery(new SpanTermQuery(new Term(FIELD_NAME, "words")), //just match the first "word" occurrence
        Collections.singletonList(new BytesRef("pos: 1")));
        IndexSearcher searcher = newSearcher(reader);
        QueryScorer scorer = new QueryScorer(query, searcher.getIndexReader(), FIELD_NAME);
        scorer.setUsePayloads(true);
        Highlighter h = new Highlighter(scorer);
        TopDocs hits = searcher.search(query, 10);
        assertEquals(1, hits.scoreDocs.length);
        TokenStream stream = TokenSources.getAnyTokenStream(searcher.getIndexReader(), 0, FIELD_NAME, analyzer);
        if (random().nextBoolean()) {
            //conceals detection of TokenStreamFromTermVector
            stream = new CachingTokenFilter(stream);
        }
        String result = h.getBestFragment(stream, text);
        //only highlight first "word"
        assertEquals("random <B>words</B> and words", result);
    }
}