Hi all, 

I'm facing some challenges, trying to understand the "Score" variable yielded on each search result by using a lucene index query. Basically what I'm trying to achieve is to implement a basic search result weight, using the Webnodes Lucene IndexQuery.

 

Webnodes versions tested: 1612, 1617.

 

 

I'm currently doing searches using my method below (included all parameters I use in case it matters).

Then, I iterate through the results filling up my own POCO class, trying to fetch the "Score" and the "ScorePercentage" variables.

I've even tried disabling fuzzy and turning on autofuzzy (q.AutoFuzzy = true), 

and disabling fuzzy entirely. 

But the "Score" variable (a float) is always "NaN", and the "ScorePercentage" (a byte) is always 0.

I use my search provider like this across several Webnodes installations, and I have not yet managed to get a >0 result on any of the search results yielded.

 

Do you have any tips as to how I can get positive score parameters on my search results?

Am I missing something? Any tips would be greatly appreciated.

 

 

private List DoWebnodesIndexSearch(string searchQuery)
        {           
            IndexQuery q = new IndexQuery();
			string fuzzySearchSimilarity = "0.8";
			if (FuzzySearch) SearchQuery = SearchQuery + "~" + fuzzySearchSimilarity; 
            if (ContentClassIdsOverride != null)
            {
                q.ClassIds = ContentClassIdsOverride;
            }
            if (SiteIdsOverride != null)
            {
                q.SiteIds = SiteIdsOverride;
            }
            q.BodySearch = searchQuery;
            q.AddSort(IndexField.ChangeDate, false);
            q.ProtectedContent = false;
            List searchResultList= WAFContext.Session.Search(q);
            return searchResultList;
        }

 

foreach (var nativeSearchResult in DoWebnodesIndexSearch(theQuery))
            {
                var someClass = new SomeClass
                {
                    ..
                    Relevance = nativeSearchResult.Score,
					RelevancePercent = nativeSearchResult.ScorePercent
					..
                };
			}

 

 

 

I can see the lucene library utilized to create scores in the Search method below.

See the TopDocs class, retrieved by indexSearcher.Search(query, null, (pageIndex + 1) * pageSize):

 

 public List Search(Query query, Sort sort, int pageIndex, int pageSize, out int pageCount, out int totalCount)

        {

            TopDocs topDoc;

            List wDocs;

            if (!WAFRuntime.Installation.EnableTextIndex)

            {

                totalCount = 0;

                pageCount = 0;

                return new List();

            }

            lock (DataAccessObject._luceneLock)

            {

                IndexSearcher indexSearcher = null;

                try

                {

                    if (pageSize == 0)

                    {

                        pageSize = 1000000;

                    }

                    indexSearcher = new IndexSearcher(this.SearchIndexer.Directory, true);

                    if (sort == null)

                    {

                        topDoc = indexSearcher.Search(query, null, (pageIndex + 1) * pageSize);

                    }

                    else

                    {

                        topDoc = indexSearcher.Search(query, null, (pageIndex + 1) * pageSize, sort);

                    }

                    ScoreDoc[] scoreDocArray = topDoc.scoreDocs;

                    List wDocs1 = new List();

                    totalCount = topDoc.totalHits;

                    pageCount = (int)Math.Ceiling((double)totalCount / (double)pageSize);

                    int num = pageIndex * pageSize;

                    int num1 = num + pageSize;

                    if (num1 > totalCount)

                    {

                        num1 = totalCount;

                    }

                    for (int i = num; i < num1; i++)

                    {

                        WDoc wDoc = new WDoc()

                        {

                            Doc = indexSearcher.Doc(scoreDocArray[i].doc),
                            Score = scoreDocArray[i].score
                        };
                        wDocs1.Add(wDoc);
                    }
                    wDocs = wDocs1;
                }
                finally
                {
                    try
                    {
                        if (indexSearcher != null)
                        {
                            indexSearcher.Close();
                        }
                    }
                    catch
                    {
                    }
                }

            }

            return wDocs;

 

Which leads me to this method in the Lucene library:

http://lucene.apache.org/core/3_0_3/api/core/org/apache/lucene/search/Searcher.html#search(org.apache.lucene.search.Query, int)

 

Should yield score results on any entry matching the query, shouldn't it?

 

 

Indeed it was.

By removing the changedate index field sort, the result yielded scores.

      //q.AddSort(IndexField.ChangeDate, false);

Thank you very much!