如何在单个搜索框同时实现Solr自动提示与拼写检查（6.3版本问题）

Hey there, let's work through the two problems you're running into with Solr 6.3 step by step.

1. StackOverflow Error with FuzzyLookupFactory as Autosuggest LookupImpl #

FuzzyLookupFactory is prone to stack overflow errors especially when dealing with large datasets or overly permissive fuzzy matching settings—this usually happens because its underlying implementation uses recursive logic to generate fuzzy matches, which can blow out the call stack if too many candidates are processed. Here's how to fix it:

• Tighten fuzzy matching constraints:
  Lower the number of allowed edits (maxEdits) to 1 (default is 2) and increase the minimum similarity threshold (fuzzyMinSimilarity) to 0.8 or 0.9. This reduces the number of recursive matches the factory needs to process.
• Switch to a more efficient lookup implementation:
  If fuzzy suggestions are still needed but you want to avoid stack issues, consider using AnalyzingInfixLookupFactory instead. It handles fuzzy logic more efficiently without relying on deep recursion, and supports both prefix and fuzzy matching.
• Validate your schema configuration:
  Make sure your suggest field's analyzer doesn't have any recursive logic (like token filters that loop indefinitely) which could compound the stack overflow issue.

Here's an example of an adjusted Suggester configuration with safer FuzzyLookupFactory settings:

<searchComponent name="suggest" class="solr.SuggestComponent">
  <lst name="suggester">
    <str name="name">mySuggester</str>
    <str name="lookupImpl">FuzzyLookupFactory</str>
    <str name="dictionaryImpl">DocumentDictionaryFactory</str>
    <str name="field">suggest_field</str>
    <float name="fuzzyMinSimilarity">0.8</float>
    <int name="maxEdits">1</int>
    <int name="maxSuggestions">10</int>
  </lst>
</searchComponent>

2. Spell Check Doesn't Return Suggestions for Correct Keywords #

This usually boils down to configuration gaps that prevent the spell checker from recognizing or suggesting valid terms. Try these fixes:

• Ensure your spell checker dictionary includes the correct term:
  If you're using an IndexBasedSpellChecker, verify that the field it's using (e.g., content) contains the keyword in your index. For FileBasedSpellChecker, check that the dictionary file has the term listed.
• Enable automatic dictionary building:
  Set buildOnCommit="true" or buildOnOptimize="true" in your spell checker configuration. This ensures the dictionary updates whenever you commit or optimize your index, so new valid terms are added automatically.
• Adjust spell check parameters:
  • Set onlyMorePopular="false"—if this is true, the spell checker will only suggest terms that are more popular than the input, which excludes the correct term itself.
  • In your query, include spellcheck=true and spellcheck.q=<your-correct-term> to explicitly trigger spell check for that term. You can also increase spellcheck.count to get more suggestions.

Example of a corrected SpellCheckComponent configuration:

<searchComponent name="spellcheck" class="solr.SpellCheckComponent">
  <lst name="spellchecker">
    <str name="name">default</str>
    <str name="classname">solr.IndexBasedSpellChecker</str>
    <str name="field">content</str>
    <str name="buildOnCommit">true</str>
    <str name="onlyMorePopular">false</str>
  </lst>
</searchComponent>

Sample query to test:
http://localhost:8983/solr/your_core/select?q=correct_term&spellcheck=true&spellcheck.q=correct_term&spellcheck.count=5

内容的提问来源于stack exchange，提问作者Sonal Patwari