Examples with DataflowException edu.uci.ics.texera.api.exception.DataflowException used on opensource projects

Example 36 with DataflowException

use of edu.uci.ics.texera.api.exception.DataflowException in project textdb by TextDB.

the class DictionaryMatcher method appendConjunctionMatchingSpans4Dictionary.

private List<Span> appendConjunctionMatchingSpans4Dictionary(Tuple inputTuple, List<String> attributeNames, List<Set<String>> queryTokenSetList, List<String> queryList) throws DataflowException {
    List<Span> matchingResults = new ArrayList<>();
    ListField<Span> payloadField = inputTuple.getField(SchemaConstants.PAYLOAD);
    List<Span> payload = payloadField.getValue();
    Map<Integer, List<Span>> relevantSpansMap = filterRelevantSpans(payload, queryTokenSetList);
    for (String attributeName : attributeNames) {
        AttributeType attributeType = inputTuple.getSchema().getAttribute(attributeName).getType();
        String fieldValue = inputTuple.getField(attributeName).getValue().toString();
        // types other than TEXT and STRING: throw Exception for now
        if (attributeType != AttributeType.STRING && attributeType != AttributeType.TEXT) {
            throw new DataflowException("KeywordMatcher: Fields other than STRING and TEXT are not supported yet");
        }
        // for STRING type, check if the dictionary entries contains the complete fieldValue
        if (attributeType == AttributeType.STRING) {
            if (queryList.contains(fieldValue)) {
                Span span = new Span(attributeName, 0, fieldValue.length(), fieldValue, fieldValue);
                matchingResults.add(span);
            }
        }
        // for TEXT type, every token in the query should be present in span
        if (attributeType == AttributeType.TEXT) {
            for (int index : relevantSpansMap.keySet()) {
                List<Span> fieldSpanList = relevantSpansMap.get(index).stream().filter(span -> span.getAttributeName().equals(attributeName)).collect(Collectors.toList());
                if (DataflowUtils.isAllQueryTokensPresent(fieldSpanList, queryTokenSetList.get(index))) {
                    matchingResults.addAll(fieldSpanList);
                }
            }
        }
    }
    return matchingResults;
}

Example 37 with DataflowException

use of edu.uci.ics.texera.api.exception.DataflowException in project textdb by TextDB.

the class FuzzyTokenMatcherSourceOperator method createLuceneQueryObject.

public static Query createLuceneQueryObject(FuzzyTokenPredicate predicate) throws DataflowException {
    try {
        /*
             * By default the boolean query takes 1024 # of clauses as the max
             * limit. Since our input query has no limitaion on the number of
             * tokens, we have to put a check.
             */
        if (predicate.getThreshold() > 1024)
            BooleanQuery.setMaxClauseCount(predicate.getThreshold() + 1);
        BooleanQuery.Builder builder = new BooleanQuery.Builder();
        builder.setMinimumNumberShouldMatch(predicate.getThreshold());
        MultiFieldQueryParser qp = new MultiFieldQueryParser(predicate.getAttributeNames().stream().toArray(String[]::new), LuceneAnalyzerConstants.getLuceneAnalyzer(predicate.getLuceneAnalyzerStr()));
        for (String s : predicate.getQueryTokens()) {
            builder.add(qp.parse(s), Occur.SHOULD);
        }
        return builder.build();
    } catch (ParseException e) {
        throw new DataflowException(e);
    }
}

Example 38 with DataflowException

use of edu.uci.ics.texera.api.exception.DataflowException in project textdb by TextDB.

the class Join method close.

@Override
public void close() throws TexeraException {
    if (cursor == CLOSED) {
        return;
    }
    try {
        innerOperator.close();
        outerOperator.close();
    } catch (Exception e) {
        throw new DataflowException(e.getMessage(), e);
    }
    // Set the inner tuple list back to null on close.
    innerTupleList = null;
    innerTupleListCursor = 0;
    cursor = CLOSED;
}

Example 39 with DataflowException

use of edu.uci.ics.texera.api.exception.DataflowException in project textdb by TextDB.

the class KeywordMatcherSourceOperator method buildConjunctionQuery.

private Query buildConjunctionQuery() throws DataflowException {
    BooleanQuery.Builder booleanQueryBuilder = new BooleanQuery.Builder();
    for (String attributeName : this.predicate.getAttributeNames()) {
        AttributeType attributeType = this.inputSchema.getAttribute(attributeName).getType();
        // types other than TEXT and STRING: throw Exception for now
        if (attributeType != AttributeType.STRING && attributeType != AttributeType.TEXT) {
            throw new DataflowException("KeywordPredicate: Fields other than STRING and TEXT are not supported yet");
        }
        if (attributeType == AttributeType.STRING) {
            Query termQuery = new TermQuery(new Term(attributeName, predicate.getQuery()));
            booleanQueryBuilder.add(termQuery, BooleanClause.Occur.SHOULD);
        }
        if (attributeType == AttributeType.TEXT) {
            BooleanQuery.Builder fieldQueryBuilder = new BooleanQuery.Builder();
            for (String token : queryTokenSet) {
                Query termQuery = new TermQuery(new Term(attributeName, token.toLowerCase()));
                fieldQueryBuilder.add(termQuery, BooleanClause.Occur.MUST);
            }
            booleanQueryBuilder.add(fieldQueryBuilder.build(), BooleanClause.Occur.SHOULD);
        }
    }
    return booleanQueryBuilder.build();
}

Example 40 with DataflowException

use of edu.uci.ics.texera.api.exception.DataflowException in project textdb by TextDB.

the class NlpSentimentOperator method open.

@Override
public void open() throws TexeraException {
    if (cursor != CLOSED) {
        return;
    }
    if (inputOperator == null) {
        throw new DataflowException(ErrorMessages.INPUT_OPERATOR_NOT_SPECIFIED);
    }
    inputOperator.open();
    Schema inputSchema = inputOperator.getOutputSchema();
    // check if input schema is present
    if (!inputSchema.containsAttribute(predicate.getInputAttributeName())) {
        throw new TexeraException(String.format("input attribute %s is not in the input schema %s", predicate.getInputAttributeName(), inputSchema.getAttributeNames()));
    }
    // check if attribute type is valid
    AttributeType inputAttributeType = inputSchema.getAttribute(predicate.getInputAttributeName()).getType();
    boolean isValidType = inputAttributeType.equals(AttributeType.STRING) || inputAttributeType.equals(AttributeType.TEXT);
    if (!isValidType) {
        throw new TexeraException(String.format("input attribute %s must have type String or Text, its actual type is %s", predicate.getInputAttributeName(), inputAttributeType));
    }
    // generate output schema by transforming the input schema
    outputSchema = transformSchema(inputOperator.getOutputSchema());
    cursor = OPENED;
    // setup NLP sentiment analysis pipeline
    Properties props = new Properties();
    props.setProperty("annotators", "tokenize, ssplit, parse, sentiment");
    sentimentPipeline = new StanfordCoreNLP(props);
}