nodeindexer.java

来自「jsr170接口的java实现。是个apache的开源项目。」· Java 代码 · 共 578 行 · 第 1/2 页

                String type = typeValue.internalValue().toString();

                // jcr:encoding is not mandatory
                String encoding = null;
                InternalValue encodingValue = getValue(QName.JCR_ENCODING);
                if (encodingValue != null) {
                    encoding = encodingValue.internalValue().toString();
                }

                InputStream stream =
                        ((BLOBFileValue) internalValue).getStream();
                Reader reader = extractor.extractText(stream, type, encoding);
                doc.add(createFulltextField(reader));
            }
        } catch (Exception e) {
            // TODO: How to recover from a transient indexing failure?
            log.warn("Exception while indexing binary property: " + e.toString());
            log.debug("Dump: ", e);
        }
    }

    /**
     * Utility method that extracts the first value of the named property
     * of the current node. Returns <code>null</code> if the property does
     * not exist or contains no values.
     *
     * @param name property name
     * @return value of the named property, or <code>null</code>
     * @throws ItemStateException if the property can not be accessed
     */
    protected InternalValue getValue(QName name) throws ItemStateException {
        try {
            PropertyId id = new PropertyId(node.getNodeId(), name);
            PropertyState property =
                (PropertyState) stateProvider.getItemState(id);
            InternalValue[] values = property.getValues();
            if (values.length > 0) {
                return values[0];
            } else {
                return null;
            }
        } catch (NoSuchItemStateException e) {
            return null;
        }
    }

    /**
     * Adds the string representation of the boolean value to the document as
     * the named field.
     *
     * @param doc           The document to which to add the field
     * @param fieldName     The name of the field to add
     * @param internalValue The value for the field to add to the document.
     */
    protected void addBooleanValue(Document doc, String fieldName, Object internalValue) {
        doc.add(new Field(FieldNames.PROPERTIES,
                FieldNames.createNamedValue(fieldName, internalValue.toString()),
                Field.Store.NO,
                Field.Index.UN_TOKENIZED,
                Field.TermVector.NO));
    }

    /**
     * Adds the calendar value to the document as the named field. The calendar
     * value is converted to an indexable string value using the {@link DateField}
     * class.
     *
     * @param doc           The document to which to add the field
     * @param fieldName     The name of the field to add
     * @param internalValue The value for the field to add to the document.
     */
    protected void addCalendarValue(Document doc, String fieldName, Object internalValue) {
        long millis = ((Calendar) internalValue).getTimeInMillis();
        doc.add(new Field(FieldNames.PROPERTIES,
                FieldNames.createNamedValue(fieldName, DateField.timeToString(millis)),
                Field.Store.NO,
                Field.Index.UN_TOKENIZED,
                Field.TermVector.NO));
    }

    /**
     * Adds the double value to the document as the named field. The double
     * value is converted to an indexable string value using the
     * {@link DoubleField} class.
     *
     * @param doc           The document to which to add the field
     * @param fieldName     The name of the field to add
     * @param internalValue The value for the field to add to the document.
     */
    protected void addDoubleValue(Document doc, String fieldName, Object internalValue) {
        double doubleVal = ((Double) internalValue).doubleValue();
        doc.add(new Field(FieldNames.PROPERTIES,
                FieldNames.createNamedValue(fieldName, DoubleField.doubleToString(doubleVal)),
                Field.Store.NO,
                Field.Index.UN_TOKENIZED,
                Field.TermVector.NO));
    }

    /**
     * Adds the long value to the document as the named field. The long
     * value is converted to an indexable string value using the {@link LongField}
     * class.
     *
     * @param doc           The document to which to add the field
     * @param fieldName     The name of the field to add
     * @param internalValue The value for the field to add to the document.
     */
    protected void addLongValue(Document doc, String fieldName, Object internalValue) {
        long longVal = ((Long) internalValue).longValue();
        doc.add(new Field(FieldNames.PROPERTIES,
                FieldNames.createNamedValue(fieldName, LongField.longToString(longVal)),
                Field.Store.NO,
                Field.Index.UN_TOKENIZED,
                Field.TermVector.NO));
    }

    /**
     * Adds the reference value to the document as the named field. The value's
     * string representation is added as the reference data. Additionally the
     * reference data is stored in the index.
     *
     * @param doc           The document to which to add the field
     * @param fieldName     The name of the field to add
     * @param internalValue The value for the field to add to the document.
     */
    protected void addReferenceValue(Document doc, String fieldName, Object internalValue) {
        String uuid = internalValue.toString();
        doc.add(new Field(FieldNames.PROPERTIES,
                FieldNames.createNamedValue(fieldName, uuid),
                Field.Store.YES, // store
                Field.Index.UN_TOKENIZED,
                Field.TermVector.NO));
    }

    /**
     * Adds the path value to the document as the named field. The path
     * value is converted to an indexable string value using the name space
     * mappings with which this class has been created.
     *
     * @param doc           The document to which to add the field
     * @param fieldName     The name of the field to add
     * @param internalValue The value for the field to add to the document.
     */
    protected void addPathValue(Document doc, String fieldName, Object internalValue) {
        Path path = (Path) internalValue;
        String pathString = path.toString();
        try {
            pathString = PathFormat.format(path, mappings);
        } catch (NoPrefixDeclaredException e) {
            // will never happen
        }
        doc.add(new Field(FieldNames.PROPERTIES,
                FieldNames.createNamedValue(fieldName, pathString),
                Field.Store.NO,
                Field.Index.UN_TOKENIZED,
                Field.TermVector.NO));
    }

    /**
     * Adds the string value to the document both as the named field and for
     * full text indexing.
     *
     * @param doc           The document to which to add the field
     * @param fieldName     The name of the field to add
     * @param internalValue The value for the field to add to the document.
     * @deprecated Use {@link #addStringValue(Document, String, Object, boolean)
     *             addStringValue(Document, String, Object, boolean)} instead.
     */
    protected void addStringValue(Document doc, String fieldName, Object internalValue) {
        addStringValue(doc, fieldName, internalValue, true);
    }

    /**
     * Adds the string value to the document both as the named field and
     * optionally for full text indexing if <code>tokenized</code> is
     * <code>true</code>.
     *
     * @param doc           The document to which to add the field
     * @param fieldName     The name of the field to add
     * @param internalValue The value for the field to add to the document.
     * @param tokenized     If <code>true</code> the string is also tokenized
     *                      and fulltext indexed.
     */
    protected void addStringValue(Document doc, String fieldName, Object internalValue, boolean tokenized) {
        String stringValue = String.valueOf(internalValue);

        // simple String
        doc.add(new Field(FieldNames.PROPERTIES,
                FieldNames.createNamedValue(fieldName, stringValue),
                Field.Store.NO,
                Field.Index.UN_TOKENIZED,
                Field.TermVector.NO));
        if (tokenized) {
            // also create fulltext index of this value
            doc.add(createFulltextField(stringValue));
            // create fulltext index on property
            int idx = fieldName.indexOf(':');
            fieldName = fieldName.substring(0, idx + 1)
                    + FieldNames.FULLTEXT_PREFIX + fieldName.substring(idx + 1);
            doc.add(new Field(fieldName, stringValue,
                    Field.Store.NO,
                    Field.Index.TOKENIZED,
                    Field.TermVector.NO));
        }
    }

    /**
     * Adds the name value to the document as the named field. The name
     * value is converted to an indexable string treating the internal value
     * as a qualified name and mapping the name space using the name space
     * mappings with which this class has been created.
     *
     * @param doc           The document to which to add the field
     * @param fieldName     The name of the field to add
     * @param internalValue The value for the field to add to the document.
     */
    protected void addNameValue(Document doc, String fieldName, Object internalValue) {
        QName qualiName = (QName) internalValue;
        String normValue = internalValue.toString();
        try {
            normValue = mappings.getPrefix(qualiName.getNamespaceURI())
                    + ":" + qualiName.getLocalName();
        } catch (NamespaceException e) {
            // will never happen
        }
        doc.add(new Field(FieldNames.PROPERTIES,
                FieldNames.createNamedValue(fieldName, normValue),
                Field.Store.NO,
                Field.Index.UN_TOKENIZED,
                Field.TermVector.NO));
    }

    /**
     * Creates a fulltext field for the string <code>value</code>.
     *
     * @param value the string value.
     * @return a lucene field.
     */
    protected Field createFulltextField(String value) {
        if (supportHighlighting) {
            // store field compressed if greater than 16k
            Field.Store stored;
            if (value.length() > 0x4000) {
                stored = Field.Store.COMPRESS;
            } else {
                stored = Field.Store.YES;
            }
            return new Field(FieldNames.FULLTEXT, value, stored,
                    Field.Index.TOKENIZED, Field.TermVector.WITH_OFFSETS);
        } else {
            return new Field(FieldNames.FULLTEXT, value,
                    Field.Store.NO, Field.Index.TOKENIZED);
        }
    }

    /**
     * Creates a fulltext field for the reader <code>value</code>.
     *
     * @param value the reader value.
     * @return a lucene field.
     */
    protected Field createFulltextField(Reader value) {
        if (supportHighlighting) {
            // need to create a string value
            StringBuffer textExtract = new StringBuffer();
            char[] buffer = new char[1024];
            int len;
            try {
                while ((len = value.read(buffer)) > -1) {
                    textExtract.append(buffer, 0, len);
                }
            } catch (IOException e) {
                log.warn("Exception reading value for fulltext field: " +
                        e.getMessage());
                log.debug("Dump:", e);
            } finally {
                try {
                    value.close();
                } catch (IOException e) {
                    // ignore
                }
            }
            return createFulltextField(textExtract.toString());
        } else {
            return new Field(FieldNames.FULLTEXT, value);
        }
    }
}