SOLR-6297: Fix for Distributed WordBreakSolrSpellChecker

git-svn-id: https://svn.apache.org/repos/asf/lucene/dev/trunk@1622476 13f79535-47bb-0310-9956-ffa450edef68

solr/CHANGES.txt

@@ -164,6 +164,9 @@ Bug Fixes
 * SOLR-6024: Fix StatsComponent when using docValues="true" multiValued="true"
   (Vitaliy Zhovtyuk & Tomas Fernandez-Lobbe via hossman)
 
+* SOLR-6297: Fix WordBreakSolrSpellChecker to not lose suggestions in shard/cloud 
+  environments (James Dyer)
+
 Other Changes
 ---------------------
 

solr/core/src/java/org/apache/solr/handler/component/SpellCheckComponent.java

@@ -192,17 +192,18 @@ public class SpellCheckComponent extends SearchComponent implements SolrCoreAwar
         boolean isCorrectlySpelled = hits > (maxResultsForSuggest==null ? 0 : maxResultsForSuggest);
         
         NamedList response = new SimpleOrderedMap();        
-        
         NamedList suggestions = toNamedList(shardRequest, spellingResult, q, extendedResults);
         response.add("suggestions", suggestions);
         
         if (extendedResults) {
           response.add("correctlySpelled", isCorrectlySpelled);
         }        
-        
         if (collate) {
           addCollationsToResponse(params, spellingResult, rb, q, response, spellChecker.isSuggestionsMayOverlap());
         }
+        if (shardRequest) {
+          addOriginalTermsToResponse(response, tokens);
+        }
         
         rb.rsp.add("spellcheck", response);
 
@@ -261,6 +262,14 @@ public class SpellCheckComponent extends SearchComponent implements SolrCoreAwar
     response.add("collations", collationList);
   }
   
+  private void addOriginalTermsToResponse(NamedList response, Collection<Token> originalTerms) {
+    List<String> originalTermStr = new ArrayList<String>();
+    for(Token t : originalTerms) {
+      originalTermStr.add(t.toString());
+    }
+    response.add("originalTerms", originalTermStr);
+  }
+
   /**
    * For every param that is of the form "spellcheck.[dictionary name].XXXX=YYYY, add
    * XXXX=YYYY as a param to the custom param list
@@ -392,8 +401,14 @@ public class SpellCheckComponent extends SearchComponent implements SolrCoreAwar
     
   @SuppressWarnings("unchecked")
   private void collectShardSuggestions(NamedList nl, SpellCheckMergeData mergeData) {
-    System.out.println(nl);
     SpellCheckResponse spellCheckResp = new SpellCheckResponse(nl);    
+    Iterable<Object> originalTermStrings = (Iterable<Object>) nl.get("originalTerms");
+    if(originalTermStrings!=null) {
+      mergeData.originalTerms = new HashSet<>();
+      for (Object originalTermObj : originalTermStrings) {
+        mergeData.originalTerms.add(originalTermObj.toString());
+      }
+    }
     for (SpellCheckResponse.Suggestion suggestion : spellCheckResp.getSuggestions()) {
       mergeData.origVsSuggestion.put(suggestion.getToken(), suggestion);
       HashSet<String> suggested = mergeData.origVsSuggested.get(suggestion.getToken());
@@ -615,8 +630,8 @@ public class SpellCheckComponent extends SearchComponent implements SolrCoreAwar
         }
         
         if (hasFreqInfo) {
-          int tokenFrequency = spellingResult.getTokenFrequency(inputToken);
+          Integer tokenFrequency = spellingResult.getTokenFrequency(inputToken);
-          if (tokenFrequency == 0) {
+          if (tokenFrequency==null || tokenFrequency == 0) {
             hasZeroFrequencyToken = true;
           }
         }

solr/core/src/java/org/apache/solr/handler/component/SpellCheckMergeData.java

@@ -17,10 +17,12 @@ package org.apache.solr.handler.component;
  * limitations under the License.
  */
 
+import java.util.Collections;
 import java.util.HashMap;
 import java.util.HashSet;
 import java.util.LinkedHashMap;
 import java.util.Map;
+import java.util.Set;
 
 import org.apache.lucene.search.spell.SuggestWord;
 import org.apache.solr.client.solrj.response.SpellCheckResponse;
@@ -39,5 +41,14 @@ public class SpellCheckMergeData {
   // alternative string -> corresponding SuggestWord object
   public Map<String, SuggestWord> suggestedVsWord = new HashMap<>();
   public Map<String, SpellCheckCollation> collations = new HashMap<>();
+  //The original terms from the user's query.
+  public Set<String> originalTerms = null;
   public int totalNumberShardResponses = 0;
+  
+  public boolean isOriginalToQuery(String term) {
+    if(originalTerms==null) {
+      return true;
+    }
+    return originalTerms.contains(term);
+  }
 }

solr/core/src/java/org/apache/solr/spelling/ConjunctionSolrSpellChecker.java

@@ -19,6 +19,7 @@ package org.apache.solr.spelling;
 
 import java.io.IOException;
 import java.util.ArrayList;
+import java.util.Collections;
 import java.util.HashMap;
 import java.util.Iterator;
 import java.util.LinkedHashMap;
@@ -167,15 +168,18 @@ public class ConjunctionSolrSpellChecker extends SolrSpellChecker {
             Map.Entry<String,Integer> corr = iter.next();
             combinedResult.add(original, corr.getKey(), corr.getValue());
             Integer tokenFrequency = combinedTokenFrequency.get(original);
-            if(tokenFrequency!=null) {
+            combinedResult.addFrequency(original, tokenFrequency==null ? 0 : tokenFrequency);
-              combinedResult.addFrequency(original, tokenFrequency);
-            }
             if(++numberAdded==numSug) {
               break;
             }
           }
         }        
         if(!anyData) {
+          if(numberAdded==0) {
+            combinedResult.add(original, Collections.<String>emptyList());
+            Integer tokenFrequency = combinedTokenFrequency.get(original);
+            combinedResult.addFrequency(original, tokenFrequency==null ? 0 : tokenFrequency);
+          }
           break;
         }        
       }      

solr/core/src/java/org/apache/solr/spelling/SolrSpellChecker.java

@@ -100,9 +100,11 @@ public abstract class SolrSpellChecker {
     for (Map.Entry<String, HashSet<String>> entry : mergeData.origVsSuggested.entrySet()) {
       String original = entry.getKey();
       
-      //Only use this suggestion if all shards reported it as misspelled.
+      //Only use this suggestion if all shards reported it as misspelled, 
+      //unless it was not a term original to the user's query
+      //(WordBreakSolrSpellChecker can add new terms to the response, and we want to keep these)
       Integer numShards = mergeData.origVsShards.get(original);
-      if(numShards<mergeData.totalNumberShardResponses) {
+      if(numShards<mergeData.totalNumberShardResponses && mergeData.isOriginalToQuery(original)) {
         continue;
       }
       

solr/core/src/java/org/apache/solr/spelling/WordBreakSolrSpellChecker.java

@@ -202,6 +202,7 @@ public class WordBreakSolrSpellChecker extends SolrSpellChecker {
     List<Term> termArr = new ArrayList<>(options.tokens.size() + 2);
     
     List<ResultEntry> breakSuggestionList = new ArrayList<>();
+    List<ResultEntry> noBreakSuggestionList = new ArrayList<>();
     boolean lastOneProhibited = false;
     boolean lastOneRequired = false;
     boolean lastOneprocedesNewBooleanOp = false;
@@ -228,6 +229,9 @@ public class WordBreakSolrSpellChecker extends SolrSpellChecker {
       if (breakWords) {
         SuggestWord[][] breakSuggestions = wbsp.suggestWordBreaks(thisTerm,
             numSuggestions, ir, options.suggestMode, sortMethod);
+        if(breakSuggestions.length==0) {
+          noBreakSuggestionList.add(new ResultEntry(tokenArr[i], null, 0));
+        }
         for (SuggestWord[] breakSuggestion : breakSuggestions) {
           sb.delete(0, sb.length());
           boolean firstOne = true;
@@ -249,6 +253,8 @@ public class WordBreakSolrSpellChecker extends SolrSpellChecker {
         }
       }
     }
+    breakSuggestionList.addAll(noBreakSuggestionList);
+    
     List<ResultEntry> combineSuggestionList = Collections.emptyList();
     CombineSuggestion[] combineSuggestions = wbsp.suggestWordCombinations(
         termArr.toArray(new Term[termArr.size()]), numSuggestions, ir, options.suggestMode);
@@ -282,33 +288,24 @@ public class WordBreakSolrSpellChecker extends SolrSpellChecker {
     int combineCount = 0;
     while (lastBreak != null || lastCombine != null) {
       if (lastBreak == null) {
-        result.add(lastCombine.token, lastCombine.suggestion, lastCombine.freq);
+        addToResult(result, lastCombine.token, getCombineFrequency(ir, lastCombine.token), lastCombine.suggestion, lastCombine.freq);
-        result.addFrequency(lastCombine.token, getCombineFrequency(ir, lastCombine.token));
         lastCombine = null;
       } else if (lastCombine == null) {
-        result.add(lastBreak.token, lastBreak.suggestion, lastBreak.freq);
+        addToResult(result, lastBreak.token, ir.docFreq(new Term(field, lastBreak.token.toString())), lastBreak.suggestion, lastBreak.freq);
-        result.addFrequency(lastBreak.token, ir.docFreq(new Term(field, lastBreak.token.toString())));
         lastBreak = null;
       } else if (lastBreak.freq < lastCombine.freq) {
-        result.add(lastCombine.token, lastCombine.suggestion, lastCombine.freq);
+        addToResult(result, lastCombine.token, getCombineFrequency(ir, lastCombine.token), lastCombine.suggestion, lastCombine.freq);        
-        result.addFrequency(lastCombine.token, getCombineFrequency(ir, lastCombine.token));
         lastCombine = null;
       } else if (lastCombine.freq < lastBreak.freq) {
-        result.add(lastBreak.token, lastBreak.suggestion, lastBreak.freq);
+        addToResult(result, lastBreak.token, ir.docFreq(new Term(field, lastBreak.token.toString())), lastBreak.suggestion, lastBreak.freq);
-        result.addFrequency(lastBreak.token, ir.docFreq(new Term(field, lastBreak.token.toString())));
         lastBreak = null;
-      } else if (breakCount >= combineCount) {
+      } else if (breakCount >= combineCount) { //TODO: Should reverse >= to < ??S
-        result.add(lastCombine.token, lastCombine.suggestion, lastCombine.freq);
+        addToResult(result, lastCombine.token, getCombineFrequency(ir, lastCombine.token), lastCombine.suggestion, lastCombine.freq);        
-        result.addFrequency(lastCombine.token, getCombineFrequency(ir, lastCombine.token));
         lastCombine = null;
       } else {
-        result.add(lastBreak.token, lastBreak.suggestion, lastBreak.freq);
+        addToResult(result, lastBreak.token, ir.docFreq(new Term(field, lastBreak.token.toString())), lastBreak.suggestion, lastBreak.freq);        
-        result.addFrequency(lastBreak.token, ir.docFreq(new Term(field, lastBreak.token.toString())));
         lastBreak = null;
       }
-      if (result.getSuggestions().size() > numSuggestions) {
-        break;
-      }
       if (lastBreak == null && breakIter.hasNext()) {
         lastBreak = breakIter.next();
         breakCount++;
@@ -320,6 +317,15 @@ public class WordBreakSolrSpellChecker extends SolrSpellChecker {
     }
     return result;
   }
+  private void addToResult(SpellingResult result, Token token, int tokenFrequency, String suggestion, int suggestionFrequency) {
+    if(suggestion==null) {
+      result.add(token, Collections.<String>emptyList());
+      result.addFrequency(token, tokenFrequency);
+    } else {
+      result.add(token, suggestion, suggestionFrequency);
+      result.addFrequency(token, tokenFrequency);
+    }
+  }
   
   private int getCombineFrequency(IndexReader ir, Token token) throws IOException {
     String[] words = spacePattern.split(token.toString());

solr/core/src/test/org/apache/solr/handler/component/DistributedSpellCheckComponentTest.java

@@ -187,6 +187,12 @@ public class DistributedSpellCheckComponentTest extends BaseDistributedSearchTes
     query(buildRequest("lowerfilt:(+quock +redfox +jum +ped)", 
         false, reqHandlerWithWordbreak, random().nextBoolean(), extended, "true", count, "10", 
         collate, "true", maxCollationTries, "0", maxCollations, "1", collateExtended, "true"));
+    query(buildRequest("lowerfilt:(+rodfix)", 
+        false, reqHandlerWithWordbreak, random().nextBoolean(), extended, "true", count, "10", 
+        collate, "true", maxCollationTries, "0", maxCollations, "1", collateExtended, "true"));
+    query(buildRequest("lowerfilt:(+son +ata)", 
+        false, reqHandlerWithWordbreak, random().nextBoolean(), extended, "true", count, "10", 
+        collate, "true", maxCollationTries, "0", maxCollations, "1", collateExtended, "true"));
   }
   private Object[] buildRequest(String q, boolean useSpellcheckQ, String handlerName, boolean useGrouping, String... addlParams) {
     List<Object> params = new ArrayList<>();

solr/core/src/test/org/apache/solr/spelling/WordBreakSolrSpellCheckerTest.java

@@ -76,7 +76,7 @@ public class WordBreakSolrSpellCheckerTest extends SolrTestCaseJ4 {
     searcher.decref();
     
     assertTrue(result != null && result.getSuggestions() != null);
-    assertTrue(result.getSuggestions().size()==6);
+    assertTrue(result.getSuggestions().size()==9);
     
     for(Map.Entry<Token, LinkedHashMap<String, Integer>> s : result.getSuggestions().entrySet()) {
       Token orig = s.getKey();
@@ -119,6 +119,27 @@ public class WordBreakSolrSpellCheckerTest extends SolrTestCaseJ4 {
         assertTrue(orig.length()==4);
         assertTrue(corr.length==1);
         assertTrue(corr[0].equals("pi ne"));
+      } else if(orig.toString().equals("pine")) {
+        assertTrue(orig.startOffset()==10);
+        assertTrue(orig.endOffset()==14);
+        assertTrue(orig.length()==4);
+        assertTrue(corr.length==1);
+        assertTrue(corr[0].equals("pi ne"));
+      } else if(orig.toString().equals("apple")) {
+        assertTrue(orig.startOffset()==15);
+        assertTrue(orig.endOffset()==20);
+        assertTrue(orig.length()==5);
+        assertTrue(corr.length==0);
+      } else if(orig.toString().equals("good")) {
+        assertTrue(orig.startOffset()==21);
+        assertTrue(orig.endOffset()==25);
+        assertTrue(orig.length()==4);
+        assertTrue(corr.length==0);
+      } else if(orig.toString().equals("ness")) {
+        assertTrue(orig.startOffset()==26);
+        assertTrue(orig.endOffset()==30);
+        assertTrue(orig.length()==4);
+        assertTrue(corr.length==0);
       }else {
         fail("Unexpected original result: " + orig);
       }