/*
* Copyright (c) Sematext International
* All Rights Reserved
*
* THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF Sematext International
* The copyright notice above does not evidence any
* actual or intended publication of such source code.
*/
package com.sematext.solr.handler.component;
import org.apache.lucene.analysis.core.WhitespaceTokenizer;
import org.apache.lucene.analysis.core.WhitespaceTokenizerFactory;
import org.apache.lucene.analysis.tokenattributes.CharTermAttribute;
import org.apache.lucene.util.AttributeFactory;
import org.apache.lucene.util.Version;
import org.apache.solr.common.util.NamedList;
import org.apache.solr.common.util.SimpleOrderedMap;
import org.apache.solr.handler.component.ResponseBuilder;
import org.apache.solr.response.SolrQueryResponse;
import java.io.IOException;
import java.io.StringReader;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.List;
import java.util.Map;
/**
* Class with useful utility methods.
*
* @author sematext, http://www.sematext.com/
*/
@SuppressWarnings({ "rawtypes" })
public class ReSearcherUtils {
public static final String INTERNAL_QUERY_MARKER_PARAM = "internalQuery";
public static long extractOriginalQueryHits(ResponseBuilder rb) {
long hits = 0;
if (rb.rsp != null) {
Number hitsInteger = (Number) rb.rsp.getToLog().get("hits");
if (hitsInteger == null) {
hits = rb.getNumberDocumentsFound();
} else {
hits = hitsInteger.longValue();
}
}
return hits;
}
/**
* Returns spellchecker's suggestions from original response. In case there are no suggestions, returns null.
*
* @param rb .
* @return .
*/
public static NamedList extractSpellcheckerSuggestions(ResponseBuilder rb) {
if (rb.rsp.getValues().get("spellcheck") == null) {
return null;
}
return (NamedList) ((SimpleOrderedMap) rb.rsp.getValues().get("spellcheck")).get("suggestions");
}
/**
* Extract spellchecker's collation from original response or null if there was no collation.
* @param rb .
* @return .
*/
public static String extractSpellcheckerCollation(ResponseBuilder rb) {
if (rb.rsp.getValues().get("spellcheck") != null) {
if (((SimpleOrderedMap) rb.rsp.getValues().get("spellcheck")).get("collations") != null) {
return (String) (((NamedList) ((SimpleOrderedMap) rb.rsp.getValues().get("spellcheck")).get("collations")).get("collation"));
}
}
return null;
}
/**
* Contains logic for iterating over spellchecker's suggestions. If any of input parameters is null, just exits. This method
* iterates over suggestions for one incorrect word (which was provided with parameter)
*
* @param suggestions list of all suggestions, can be extracted with method .extractSpellcheckerSuggestions().
* @param word term for which suggestions are being processed
* @param processor instance of processor which will handle all suggestions for word
*/
public static void iterateOverSpellcheckerSuggestionsForWord(NamedList suggestions, String word, SpellcheckerSuggestionProcessor processor) {
if (suggestions == null || word == null || processor == null) {
return;
}
NamedList stuff = (NamedList) suggestions.get(word);
iterateOverSpellcheckerSuggestionsForWord(stuff, processor);
}
/**
* Contains logic for iterating over spellchecker's suggestions. If any of input parameters is null, just exits. This method
* iterates over suggestions for one incorrect word (for which parameter <code>suggestions</code> should contain spellchecker's suggestions).
*
*
* @param suggestions list of suggestions for some word, so word parameter isn't needed
* @param processor instance of processor which will handle all suggestions for word
*/
@SuppressWarnings("unchecked")
public static void iterateOverSpellcheckerSuggestionsForWord(NamedList suggestions, SpellcheckerSuggestionProcessor processor) {
int i = 4;
// double topFreq = 0;
if (suggestions.getVal(4) instanceof List) {
// support for new version of Solr (valid after 2009-09-09 in version 1.4)
List<SimpleOrderedMap> l = (List<SimpleOrderedMap>) suggestions.getVal(4);
i = 0;
while (true) {
if (l.size() <= i) {
break;
}
processor.process(l.get(i), (String) l.get(i).get("word"));
i++;
}
}
else {
// old way, before 2009-09-09
while (true) {
if (suggestions.size() <= i) {
break;
}
processor.process((NamedList) (suggestions).getVal(i), suggestions.getName(i));
i++;
}
}
processor.afterProcessingFinished();
}
/**
* Contains logic for iterating over spellchecker's suggestions. If any of input parameters is null, just exits. This method
* iterates over suggestions for ALL incorrectly spelled words. It will not sub-iterate over suggestions of each incorrect
* word, that is the logic which SpellcheckerSuggestionProcessor should take care off.
*
* @param suggestions list of all suggestions, can be extracted with method .extractSpellcheckerSuggestions().
* @param processor instance of processor which will handle all suggestions for word
*/
@SuppressWarnings("unchecked")
public static void iterateOverSpellcheckerSuggestionsForAllIncorrectWords(NamedList suggestions, SpellcheckerSuggestionProcessor processor) {
if (suggestions == null || processor == null) {
return;
}
for (int i = 0; i < suggestions.size(); i++) {
// find first acceptable suggestion
if (suggestions.getVal(i) instanceof SimpleOrderedMap) {
processor.process((SimpleOrderedMap) suggestions.getVal(i), suggestions.getName(i));
}
}
processor.afterProcessingFinished();
}
/**
* Adds some values (most likely from (List) to response which will be sent to the client. The values will be added into
* root level under tag name specified by tagName
*
* @param rb .
* @param tagName .
* @param value .
*/
public static void addValueToResult(ResponseBuilder rb, String tagName, Object value) {
rb.rsp.add(tagName, value);
}
/**
* Utility method for creating map which should be passed into ReSearcherUtils.customPrepare as parametersToReplace when
* only q value is changed. Other similar methods will not be provided, since changed q is the most used situation.
*
* @param newQueryValue .
* @return .
*/
public static Map<String, List<String>> createParametersToReplaceMapWhenOnlyQueryChanged(String newQueryValue) {
Map<String, List<String>> parametersToReplace = new HashMap<String, List<String>>();
List<String> newQuery = new ArrayList<String>();
newQuery.add(newQueryValue);
parametersToReplace.put("q", newQuery);
return parametersToReplace;
}
/**
* Utility method for creating map which should be passed into ReSearcherUtils.customPrepare .
*
* @param str .
* @return .
*/
public static Map<String, List<String>> createParameters(String... str) {
Map<String, List<String>> parameters = new HashMap<String, List<String>>();
for(int i = 0; i < str.length / 2; i++) {
List<String> newQuery = new ArrayList<String>();
newQuery.add(str[i*2+1]);
parameters.put(str[i*2], newQuery);
}
return parameters;
}
/**
* Separates tokens from query. Treats each quote as a separate token, since that makes it easier to examine the query.
*
* @param queryString .
* @param tokens .
* @return number of quotes in the query
*/
public static int tokenizeQueryString(String queryString, List<String> tokens) {
int countOfQuotes = 0;
try {
// first tokenize words and treat each quote as a separate token
Map<String,String> args = new HashMap<String, String>();
args.put(WhitespaceTokenizerFactory.LUCENE_MATCH_VERSION_PARAM, Version.LUCENE_6_3_0.toString());
WhitespaceTokenizerFactory f = new WhitespaceTokenizerFactory(args);
WhitespaceTokenizer s = (WhitespaceTokenizer)f.create(AttributeFactory.DEFAULT_ATTRIBUTE_FACTORY);
s.setReader(new StringReader(queryString));
s.reset();
while (true) {
CharTermAttribute t = s.getAttribute(CharTermAttribute.class);
if (t == null) {
break;
}
String tokentText = new String(t.toString());
if (tokentText.equals("\"")) {
tokens.add("\"");
countOfQuotes++;
} else if (tokentText.startsWith("\"")) {
tokens.add("\"");
countOfQuotes++;
if (tokentText.endsWith("\"")) {
tokens.add(tokentText.substring(1, tokentText.length() - 1));
tokens.add("\"");
countOfQuotes++;
} else {
tokens.add(tokentText.substring(1));
}
} else if (tokentText.endsWith("\"")) {
tokens.add(tokentText.substring(0, tokentText.length() - 1));
tokens.add("\"");
countOfQuotes++;
} else if (!tokentText.trim().equals("")) {
// take into account only if different than empty string
tokens.add(tokentText);
}
if (!s.incrementToken()) {
break;
}
}
s.end();
s.close();
} catch (IOException e) {
throw new RuntimeException(e);
}
return countOfQuotes;
}
public static void addComponentResponsesToResult(SolrQueryResponse rsp, SolrQueryResponse newRsp, String prefix) {
rsp.add(prefix + "_response", newRsp.getValues().get("response"));
rsp.add(prefix + "_grouped", newRsp.getValues().get("grouped"));
rsp.add(prefix + "_facet_counts", newRsp.getValues().get("facet_counts"));
rsp.add(prefix + "_terms", newRsp.getValues().get("terms"));
rsp.add(prefix + "_termVectors", newRsp.getValues().get("termVectors"));
rsp.add(prefix + "_highlighting", newRsp.getValues().get("highlighting"));
rsp.add(prefix + "_stats", newRsp.getValues().get("stats"));
}
}
