// Handle additional arguments...
protected void setArgs(IndexSchema schema, Map<String, String> args) {
// default to STORED, INDEXED, OMIT_TF_POSITIONS and MULTIVALUED depending on schema version
properties = (STORED | INDEXED);
float schemaVersion = schema.getVersion();
if (schemaVersion < 1.1f) properties |= MULTIVALUED;
if (schemaVersion > 1.1f) properties |= OMIT_TF_POSITIONS;
if (schemaVersion < 1.3) {
args.remove("compressThreshold");
}
if (schemaVersion >= 1.6f) properties |= USE_DOCVALUES_AS_STORED;
this.args = Collections.unmodifiableMap(args);
Map<String, String> initArgs = new HashMap<>(args);
initArgs.remove(CLASS_NAME); // consume the class arg
trueProperties = FieldProperties.parseProperties(initArgs, true, false);
falseProperties = FieldProperties.parseProperties(initArgs, false, false);
properties &= ~falseProperties;
properties |= trueProperties;
for (String prop : FieldProperties.propertyNames) initArgs.remove(prop);
init(schema, initArgs);
String positionInc = initArgs.get(POSITION_INCREMENT_GAP);
if (positionInc != null) {
Analyzer analyzer = getIndexAnalyzer();
if (analyzer instanceof SolrAnalyzer) {
((SolrAnalyzer) analyzer).setPositionIncrementGap(Integer.parseInt(positionInc));
} else {
throw new RuntimeException(
"Can't set " + POSITION_INCREMENT_GAP + " on custom analyzer " + analyzer.getClass());
}
analyzer = getQueryAnalyzer();
if (analyzer instanceof SolrAnalyzer) {
((SolrAnalyzer) analyzer).setPositionIncrementGap(Integer.parseInt(positionInc));
} else {
throw new RuntimeException(
"Can't set " + POSITION_INCREMENT_GAP + " on custom analyzer " + analyzer.getClass());
}
initArgs.remove(POSITION_INCREMENT_GAP);
}
this.postingsFormat = initArgs.remove(POSTINGS_FORMAT);
this.docValuesFormat = initArgs.remove(DOC_VALUES_FORMAT);
if (initArgs.size() > 0) {
throw new RuntimeException(
"schema fieldtype "
+ typeName
+ "("
+ this.getClass().getName()
+ ")"
+ " invalid arguments:"
+ initArgs);
}
}
/**
* Begins the indexing
*
* @exception BuildException If an error occurs indexing the fileset
*/
@Override
public void execute() throws BuildException {
// construct handler and analyzer dynamically
try {
handler = Class.forName(handlerClassName).asSubclass(DocumentHandler.class).newInstance();
analyzer = IndexTask.createAnalyzer(analyzerClassName);
} catch (Exception e) {
throw new BuildException(e);
}
log("Document handler = " + handler.getClass(), Project.MSG_VERBOSE);
log("Analyzer = " + analyzer.getClass(), Project.MSG_VERBOSE);
if (handler instanceof ConfigurableDocumentHandler) {
((ConfigurableDocumentHandler) handler).configure(handlerConfig.getProperties());
}
try {
indexDocs();
} catch (IOException e) {
throw new BuildException(e);
}
}
public void testRangeQuery() throws ParseException {
for (int i = 0; i < rangeInput.length; i++) {
assertEquals(
"Testing ranges with analyzer " + a.getClass() + ", input string: " + rangeInput[i],
rangeExpected[i],
parseWithAnalyzingQueryParser(rangeInput[i], a));
}
}
public void testPrefixQuery() throws ParseException {
for (int i = 0; i < prefixInput.length; i++) {
assertEquals(
"Testing prefixes with analyzer " + a.getClass() + ", input string: " + prefixInput[i],
prefixExpected[i],
parseWithAnalyzingQueryParser(prefixInput[i], a));
}
}
public void testWildCardQuery() throws ParseException {
for (int i = 0; i < wildcardInput.length; i++) {
assertEquals(
"Testing wildcards with analyzer " + a.getClass() + ", input string: " + wildcardInput[i],
wildcardExpected[i],
parseWithAnalyzingQueryParser(wildcardInput[i], a));
}
}
public void testFuzzyQuery() throws ParseException {
for (int i = 0; i < fuzzyInput.length; i++) {
assertEquals(
"Testing fuzzys with analyzer " + a.getClass() + ", input string: " + fuzzyInput[i],
fuzzyExpected[i],
parseWithAnalyzingQueryParser(fuzzyInput[i], a));
}
}
private static String[] groupTokens(Analyzer analyzer, String input) throws IOException {
if (Resources.debug) {
Resources.LOGGER.debug("TokenParser:" + input);
Resources.LOGGER.debug("Analyzer:" + analyzer.getClass());
}
TokenStream tokenStream = analyzer.tokenStream("input", input);
OffsetAttribute offsetAttribute = tokenStream.addAttribute(OffsetAttribute.class);
PositionIncrementAttribute positionIncrementAttribute =
tokenStream.addAttribute(PositionIncrementAttribute.class);
CharTermAttribute charTermAttribute = tokenStream.addAttribute(CharTermAttribute.class);
TypeAttribute typeAttribute = tokenStream.addAttribute(TypeAttribute.class);
tokenStream.reset();
int position = 0;
List<TermInfo> infos = new ArrayList<TermInfo>();
while (tokenStream.incrementToken()) {
int increment = positionIncrementAttribute.getPositionIncrement();
if (increment > 0) {
position = position + increment;
if (Resources.debug) {
Resources.LOGGER.debug(position + ":");
}
}
int startOffset = offsetAttribute.startOffset();
int endOffset = offsetAttribute.endOffset();
String term = charTermAttribute.toString();
TermInfo info = new TermInfo();
info.setStart(startOffset);
info.setEnd(endOffset);
infos.add(info);
if (Resources.debug) {
Resources.LOGGER.debug(
"["
+ term
+ "]"
+ ":("
+ startOffset
+ "-->"
+ endOffset
+ "):"
+ typeAttribute.type());
}
}
tokenStream.end();
tokenStream.close();
Stack<TermInfo> tiStack = groupTokenInfos(infos);
List<String> terms = new ArrayList<String>();
while (!tiStack.isEmpty()) {
TermInfo termInfo = tiStack.pop();
if (termInfo.getEnd() <= input.length() && termInfo.getStart() >= 1) {
String term = input.substring(termInfo.getStart(), termInfo.getEnd());
terms.add(term);
}
}
return terms.toArray(new String[] {});
}
@Override
public int doLogic() throws Exception {
try {
Locale locale = getRunData().getLocale();
if (locale == null) throw new RuntimeException("Locale must be set with the NewLocale task!");
Analyzer analyzer = createAnalyzer(locale, impl);
getRunData().setAnalyzer(analyzer);
System.out.println(
"Changed Analyzer to: " + analyzer.getClass().getName() + "(" + locale + ")");
} catch (Exception e) {
throw new RuntimeException("Error creating Analyzer: impl=" + impl, e);
}
return 1;
}
private static String[] mmsegTokens(Analyzer analyzer, String input) throws IOException {
if (Resources.debug) {
Resources.LOGGER.debug("TokenParser:" + input);
Resources.LOGGER.debug("Analyzer:" + analyzer.getClass());
}
TokenStream tokenStream = analyzer.tokenStream("input", input);
OffsetAttribute offsetAttribute = tokenStream.addAttribute(OffsetAttribute.class);
PositionIncrementAttribute positionIncrementAttribute =
tokenStream.addAttribute(PositionIncrementAttribute.class);
CharTermAttribute charTermAttribute = tokenStream.addAttribute(CharTermAttribute.class);
TypeAttribute typeAttribute = tokenStream.addAttribute(TypeAttribute.class);
tokenStream.reset();
int position = 0;
List<String> tokens = new ArrayList<String>();
while (tokenStream.incrementToken()) {
int increment = positionIncrementAttribute.getPositionIncrement();
if (increment > 0) {
position = position + increment;
if (Resources.debug) {
Resources.LOGGER.debug(position + ":");
}
}
int startOffset = offsetAttribute.startOffset();
int endOffset = offsetAttribute.endOffset();
String term = charTermAttribute.toString();
tokens.add(term);
if (Resources.debug) {
Resources.LOGGER.debug(
"["
+ term
+ "]"
+ ":("
+ startOffset
+ "-->"
+ endOffset
+ "):"
+ typeAttribute.type());
}
}
tokenStream.end();
tokenStream.close();
return tokens.toArray(new String[] {});
}
/* (non-Javadoc)
* @see org.apache.hadoop.mapred.MapReduceBase#configure(org.apache.hadoop.mapred.JobConf)
*/
public void configure(JobConf job) {
iconf = new IndexUpdateConfiguration(job);
analyzer = (Analyzer) ReflectionUtils.newInstance(iconf.getDocumentAnalyzerClass(), job);
localAnalysis =
(ILocalAnalysis) ReflectionUtils.newInstance(iconf.getLocalAnalysisClass(), job);
localAnalysis.configure(job);
shards = Shard.getIndexShards(iconf);
distributionPolicy =
(IDistributionPolicy) ReflectionUtils.newInstance(iconf.getDistributionPolicyClass(), job);
distributionPolicy.init(shards);
LOG.info("sea.document.analyzer = " + analyzer.getClass().getName());
LOG.info("sea.local.analysis = " + localAnalysis.getClass().getName());
LOG.info(shards.length + " shards = " + iconf.getIndexShards());
LOG.info("sea.distribution.policy = " + distributionPolicy.getClass().getName());
}
private static SimpleOrderedMap<Object> getAnalyzerInfo(Analyzer analyzer) {
SimpleOrderedMap<Object> aninfo = new SimpleOrderedMap<Object>();
aninfo.add("className", analyzer.getClass().getName());
if (analyzer instanceof TokenizerChain) {
TokenizerChain tchain = (TokenizerChain) analyzer;
CharFilterFactory[] cfiltfacs = tchain.getCharFilterFactories();
SimpleOrderedMap<Map<String, Object>> cfilters = new SimpleOrderedMap<Map<String, Object>>();
for (CharFilterFactory cfiltfac : cfiltfacs) {
Map<String, Object> tok = new HashMap<String, Object>();
String className = cfiltfac.getClass().getName();
tok.put("className", className);
tok.put("args", cfiltfac.getArgs());
cfilters.add(className.substring(className.lastIndexOf('.') + 1), tok);
}
if (cfilters.size() > 0) {
aninfo.add("charFilters", cfilters);
}
SimpleOrderedMap<Object> tokenizer = new SimpleOrderedMap<Object>();
TokenizerFactory tfac = tchain.getTokenizerFactory();
tokenizer.add("className", tfac.getClass().getName());
tokenizer.add("args", tfac.getArgs());
aninfo.add("tokenizer", tokenizer);
TokenFilterFactory[] filtfacs = tchain.getTokenFilterFactories();
SimpleOrderedMap<Map<String, Object>> filters = new SimpleOrderedMap<Map<String, Object>>();
for (TokenFilterFactory filtfac : filtfacs) {
Map<String, Object> tok = new HashMap<String, Object>();
String className = filtfac.getClass().getName();
tok.put("className", className);
tok.put("args", filtfac.getArgs());
filters.add(className.substring(className.lastIndexOf('.') + 1), tok);
}
if (filters.size() > 0) {
aninfo.add("filters", filters);
}
}
return aninfo;
}
public void analyze(String text) throws IOException {
List<String> searchlst = new ArrayList<String>();
proposalController.getProposalList().clear();
String query = "";
System.out.println("Analzying \"" + text + "\"");
Analyzer analyzer = new RussianAnalyzer(Version.LUCENE_31);
System.out.println("\t" + analyzer.getClass().getName() + ":");
System.out.print("\t\t");
TokenStream stream = analyzer.tokenStream("contents", new StringReader(text));
while (true) {
if (!stream.incrementToken()) break;
AttributeSource token = stream.cloneAttributes();
CharTermAttribute term = (CharTermAttribute) token.addAttribute(CharTermAttribute.class);
System.out.print("[" + term.toString() + "] "); // 2
searchlst.add(term.toString());
}
int i = 0;
for (String param : searchlst) {
if (i < searchlst.size() - 1) {
query += param + " AND ";
} else {
query += param;
}
i++;
}
_log.info("Запрос для поиска:" + query);
startSearch(query);
System.out.println("\n");
}
/**
* Returns a description of the given analyzer, by either reporting the Analyzer class name (and
* optionally luceneMatchVersion) if it's not a TokenizerChain, or if it is, querying each
* analysis factory for its name and args.
*/
protected static SimpleOrderedMap<Object> getAnalyzerProperties(Analyzer analyzer) {
SimpleOrderedMap<Object> analyzerProps = new SimpleOrderedMap<>();
if (analyzer instanceof TokenizerChain) {
Map<String, String> factoryArgs;
TokenizerChain tokenizerChain = (TokenizerChain) analyzer;
CharFilterFactory[] charFilterFactories = tokenizerChain.getCharFilterFactories();
if (0 < charFilterFactories.length) {
List<SimpleOrderedMap<Object>> charFilterProps = new ArrayList<>();
for (CharFilterFactory charFilterFactory : charFilterFactories) {
SimpleOrderedMap<Object> props = new SimpleOrderedMap<>();
props.add(CLASS_NAME, charFilterFactory.getClassArg());
factoryArgs = charFilterFactory.getOriginalArgs();
if (null != factoryArgs) {
for (String key : factoryArgs.keySet()) {
if (!CLASS_NAME.equals(key)) {
if (LUCENE_MATCH_VERSION_PARAM.equals(key)) {
if (charFilterFactory.isExplicitLuceneMatchVersion()) {
props.add(key, factoryArgs.get(key));
}
} else {
props.add(key, factoryArgs.get(key));
}
}
}
}
charFilterProps.add(props);
}
analyzerProps.add(CHAR_FILTERS, charFilterProps);
}
SimpleOrderedMap<Object> tokenizerProps = new SimpleOrderedMap<>();
TokenizerFactory tokenizerFactory = tokenizerChain.getTokenizerFactory();
tokenizerProps.add(CLASS_NAME, tokenizerFactory.getClassArg());
factoryArgs = tokenizerFactory.getOriginalArgs();
if (null != factoryArgs) {
for (String key : factoryArgs.keySet()) {
if (!CLASS_NAME.equals(key)) {
if (LUCENE_MATCH_VERSION_PARAM.equals(key)) {
if (tokenizerFactory.isExplicitLuceneMatchVersion()) {
tokenizerProps.add(key, factoryArgs.get(key));
}
} else {
tokenizerProps.add(key, factoryArgs.get(key));
}
}
}
}
analyzerProps.add(TOKENIZER, tokenizerProps);
TokenFilterFactory[] filterFactories = tokenizerChain.getTokenFilterFactories();
if (0 < filterFactories.length) {
List<SimpleOrderedMap<Object>> filterProps = new ArrayList<>();
for (TokenFilterFactory filterFactory : filterFactories) {
SimpleOrderedMap<Object> props = new SimpleOrderedMap<>();
props.add(CLASS_NAME, filterFactory.getClassArg());
factoryArgs = filterFactory.getOriginalArgs();
if (null != factoryArgs) {
for (String key : factoryArgs.keySet()) {
if (!CLASS_NAME.equals(key)) {
if (LUCENE_MATCH_VERSION_PARAM.equals(key)) {
if (filterFactory.isExplicitLuceneMatchVersion()) {
props.add(key, factoryArgs.get(key));
}
} else {
props.add(key, factoryArgs.get(key));
}
}
}
}
filterProps.add(props);
}
analyzerProps.add(FILTERS, filterProps);
}
} else { // analyzer is not instanceof TokenizerChain
analyzerProps.add(CLASS_NAME, analyzer.getClass().getName());
if (analyzer.getVersion() != Version.LATEST) {
analyzerProps.add(LUCENE_MATCH_VERSION_PARAM, analyzer.getVersion().toString());
}
}
return analyzerProps;
}
