SOLR-12089: FileBasedSpellChecker docs have some missing params (#2356)

* Now handles a accept accuracy as float.   
* deprecate misspelled `breakSugestionTieBreaker` parameter in favor of `breakSuggestionTieBreaker`
  in WordBreakSolrSpellChecker.
* Audit and update the Ref Guide for missing parameters.

---------

Co-authored-by: Andrey Bozhko <[email protected]>
Co-authored-by: Eric Pugh <[email protected]>
Co-authored-by: Christine Poerschke <[email protected]>

Author: 4 people

solr/CHANGES.txt

@@ -162,6 +162,10 @@ Other Changes
 
 * SOLR-17190: Replace org.apache.solr.util.LongSet with hppc LongHashSet (Michael Gibney)
 
+* SOLR-12089: Update FileBasedSpellChecker and IndexBasedSpellChecker to accept accuracy parameter
+  as float; deprecate `breakSugestionTieBreaker` parameter in favor of `breakSuggestionTieBreaker`
+  in WordBreakSolrSpellChecker (Andrey Bozhko via Eric Pugh)
+
 * SOLR-17201: Http2SolrClient and friends no longer marked as @lucene.experimental.
   Krb5HttpClientBuilder and PreemptiveBasicAuthClientBuilderFactory no longer deprecated (janhoy)
 

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

@@ -760,7 +760,7 @@ public void inform(SolrCore core) {
   private boolean addSpellChecker(SolrCore core, boolean hasDefault, NamedList<?> spellchecker) {
     String className = (String) spellchecker.get("classname");
     if (className == null) className = (String) spellchecker.get("class");
-    // TODO: this is a little bit sneaky: warn if class isnt supplied
+    // TODO: this is a little bit sneaky: warn if class isn't supplied
     // so that it's mandatory in a future release?
     if (className == null) className = IndexBasedSpellChecker.class.getName();
     SolrResourceLoader loader = core.getResourceLoader();

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

@@ -81,7 +81,6 @@ public abstract class AbstractLuceneSpellChecker extends SolrSpellChecker {
   public String init(NamedList<?> config, SolrCore core) {
     super.init(config, core);
     indexDir = (String) config.get(INDEX_DIR);
-    String accuracy = (String) config.get(ACCURACY);
     // If indexDir is relative then create index inside core.getDataDir()
     if (indexDir != null) {
       if (!new File(indexDir).isAbsolute()) {
@@ -120,9 +119,10 @@ public String init(NamedList<?> config, SolrCore core) {
     } catch (IOException e) {
       throw new RuntimeException(e);
     }
+    Object accuracy = config.get(ACCURACY);
     if (accuracy != null) {
       try {
-        this.accuracy = Float.parseFloat(accuracy);
+        this.accuracy = Float.parseFloat(accuracy.toString());
         spellChecker.setAccuracy(this.accuracy);
       } catch (NumberFormatException e) {
         throw new RuntimeException("Unparseable accuracy given for dictionary: " + name, e);

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

@@ -17,6 +17,7 @@
 package org.apache.solr.spelling;
 
 import java.io.IOException;
+import java.lang.invoke.MethodHandles;
 import java.util.ArrayList;
 import java.util.Collections;
 import java.util.Iterator;
@@ -32,6 +33,8 @@
 import org.apache.solr.common.util.NamedList;
 import org.apache.solr.core.SolrCore;
 import org.apache.solr.search.SolrIndexSearcher;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
 
 /**
  * A spellchecker that breaks and combines words.
@@ -46,6 +49,9 @@
  * properly sets these flags.
  */
 public class WordBreakSolrSpellChecker extends SolrSpellChecker {
+
+  private static final Logger log = LoggerFactory.getLogger(MethodHandles.lookup().lookupClass());
+
   /** Try to combine multiple words into one? [true|false] */
   public static final String PARAM_COMBINE_WORDS = "combineWords";
 
@@ -61,16 +67,24 @@ public class WordBreakSolrSpellChecker extends SolrSpellChecker {
   /** See {@link WordBreakSpellChecker#setMinBreakWordLength} */
   public static final String PARAM_MIN_BREAK_WORD_LENGTH = "minBreakLength";
 
+  /**
+   * See {@link BreakSuggestionTieBreaker} for options.
+   *
+   * @deprecated Only used for backwards compatibility. It will be removed in 10.x.
+   */
+  @Deprecated(since = "9.6")
+  private static final String PARAM_BREAK_SUGESTION_TIE_BREAKER = "breakSugestionTieBreaker";
+
   /** See {@link BreakSuggestionTieBreaker} for options. */
-  public static final String PARAM_BREAK_SUGGESTION_TIE_BREAKER = "breakSugestionTieBreaker";
+  public static final String PARAM_BREAK_SUGGESTION_TIE_BREAKER = "breakSuggestionTieBreaker";
 
   /** See {@link WordBreakSpellChecker#setMaxEvaluations} */
   public static final String PARAM_MAX_EVALUATIONS = "maxEvaluations";
 
   /** See {@link WordBreakSpellChecker#setMinSuggestionFrequency} */
   public static final String PARAM_MIN_SUGGESTION_FREQUENCY = "minSuggestionFreq";
 
-  /** Specify a value on the "breakSugestionTieBreaker" parameter. The default is MAX_FREQ. */
+  /** Specify a value on the "breakSuggestionTieBreaker" parameter. The default is MAX_FREQ. */
   public enum BreakSuggestionTieBreaker {
     /** See {@link BreakSuggestionSortMethod#NUM_CHANGES_THEN_MAX_FREQUENCY} # */
     MAX_FREQ,
@@ -92,6 +106,17 @@ public String init(NamedList<?> config, SolrCore core) {
     breakWords = boolParam(config, PARAM_BREAK_WORDS);
     wbsp = new WordBreakSpellChecker();
     String bstb = strParam(config, PARAM_BREAK_SUGGESTION_TIE_BREAKER);
+    if (bstb == null) {
+      bstb = strParam(config, PARAM_BREAK_SUGESTION_TIE_BREAKER);
+      if (bstb != null && log.isWarnEnabled()) {
+        log.warn(
+            "Parameter '"
+                + PARAM_BREAK_SUGESTION_TIE_BREAKER
+                + "' is deprecated and will be removed in Solr 10.x. Please use '"
+                + PARAM_BREAK_SUGGESTION_TIE_BREAKER
+                + "' instead."); // nowarn
+      }
+    }
     if (bstb != null) {
       bstb = bstb.toUpperCase(Locale.ROOT);
       if (bstb.equals(BreakSuggestionTieBreaker.SUM_FREQ.name())) {

solr/core/src/test-files/solr/collection1/conf/solrconfig-minhash.xml

@@ -271,7 +271,7 @@
       <str name="name">freq</str>
       <str name="field">lowerfilt</str>
       <str name="spellcheckIndexDir">spellcheckerFreq</str>
-      <!-- comparatorClass be one of:
+      <!-- comparatorClass can be one of:
         1. score (default)
         2. freq (Frequency first, then score)
         3. A fully qualified class name

solr/core/src/test-files/solr/collection1/conf/solrconfig-plugcollector.xml

@@ -259,7 +259,7 @@
       <str name="name">freq</str>
       <str name="field">lowerfilt</str>
       <str name="spellcheckIndexDir">spellcheckerFreq</str>
-      <!-- comparatorClass be one of:
+      <!-- comparatorClass can be one of:
         1. score (default)
         2. freq (Frequency first, then score)
         3. A fully qualified class name

solr/core/src/test-files/solr/collection1/conf/solrconfig-spellcheckcomponent.xml

@@ -83,7 +83,7 @@
       <str name="field">lowerfilt</str>
       <str name="combineWords">true</str>
       <str name="breakWords">true</str>
-      <str name="breakSugestionTieBreaker">MAX_FREQ</str>
+      <str name="breakSuggestionTieBreaker">MAX_FREQ</str>
       <int name="maxChanges">10</int>
     </lst>
     <lst name="spellchecker">
@@ -122,13 +122,14 @@
       <str name="sourceLocation">spellings.txt</str>
       <str name="characterEncoding">UTF-8</str>
       <str name="spellcheckIndexDir">spellchecker3</str>
+      <float name="accuracy">0.5</float>
     </lst>
     <!-- Comparator -->
     <lst name="spellchecker">
       <str name="name">freq</str>
       <str name="field">lowerfilt</str>
       <str name="spellcheckIndexDir">spellcheckerFreq</str>
-      <!-- comparatorClass be one of:
+      <!-- comparatorClass can be one of:
         1. score (default)
         2. freq (Frequency first, then score)
         3. A fully qualified class name

solr/core/src/test-files/solr/collection1/conf/solrconfig.xml

@@ -278,7 +278,7 @@
       <str name="name">freq</str>
       <str name="field">lowerfilt</str>
       <str name="spellcheckIndexDir">spellcheckerFreq</str>
-      <!-- comparatorClass be one of:
+      <!-- comparatorClass can be one of:
         1. score (default)
         2. freq (Frequency first, then score)
         3. A fully qualified class name

solr/server/solr/configsets/sample_techproducts_configs/conf/solrconfig.xml

@@ -828,7 +828,7 @@
 
     <!-- a spellchecker that use an alternate comparator
 
-         comparatorClass be one of:
+         comparatorClass can be one of:
           1. score (default)
           2. freq (Frequency first, then score)
           3. A fully qualified class name

solr/solr-ref-guide/modules/query-guide/pages/function-queries.adoc

@@ -473,7 +473,7 @@ Uses the Lucene spell checker `StringDistance` interface and supports all of the
 Possible values for distance measure are:
 
 * jw: Jaro-Winkler
-* edit: Levenstein or Edit distance
+* edit: Levenshtein or Edit distance
 * ngram: The NGramDistance, if specified, can optionally pass in the ngram size too.
 Default is 2.
 * FQN: Fully Qualified class Name for an implementation of the StringDistance interface.