Text Formatting/Highlighting in the Concordance Results window.

Question

I am unable to locate a reference from the SDL SDK that helps me understand how to apply formatting on the (TranslationUnit) text that I send to the Concordance (and/or Lookup) result windows; can you please advise how to do this with an example? 
 I am assuming that I need to encapsulate the text/words that I want highlighted with a tag reference; after a day of routing around with this, I am not able to understand what tag reference is to be used to accomplish this from the TranslationUnit / SegmentElements. I have completed development of returning TranslationUnits to the Concordance (and Lookup) result windows in Studio, but I would also like to format the individual words, as seen from the attached snapshot 
 Example : If I do a concordance search for the word &ldquo; Location &rdquo;, then I would like to format the word &ldquo; Location &rdquo; with a yellow background in TranslationUnit Source that is returned to the Concordance result window. Make reference to the attached snapshot.

Note : I am using the example &ldquo; Translation Memory Provider example &rdquo; that is provided with the SDK 2.0 to try understand this. From this example provided by SDL, the results are returned in a TranslationUnit; I am assuming that it is possible to apply the font formatting on the SegmentElements in the TranslationUnit.SourceSegment, so that highlights/formats the words accordingly that are returned in the Concordance (and/or Lookup) result windows in Studio. 
 
 Thank you for your support, 
 Patrick.

Xingzeng Liu · Accepted Answer

The returned SearchResult.Tokens can provide word information about the segment, and searchResult.ScoringResult.MatchingConcordanceRanges can give information where the concordance matches. These information could be used in Concordance view pane to highlight the matches. Here I created a simple example to demonstrate how this works.

I just modify the concordance search implementation in file ListTranslationProviderLanguageDirection.cs as the following:

public SearchResults SearchText(SearchSettings settings, string searchString)
{
    Segment s = new Sdl.LanguagePlatform.Core.Segment(_languageDirection.SourceCulture);
    s.Add(searchString);
    var searchResults = SearchSegment(settings, s);
    foreach (var searchResult in searchResults)
    {
        var segment = searchResult.MemoryTranslationUnit.SourceSegment;
        segment.Tokens = Tokenize(segment);
        searchResult.ScoringResult.MatchingConcordanceRanges = CollectConcordanceMatchRanges(segment, searchString);
    }
    return searchResults;
}

// a naive tokenizer implementation to treat a single text run as one token.
private List<Token> Tokenize(Segment segment)
{
    var tokens = new List<Token>();
    int run = 0;
    foreach (var element in segment.Elements)
    {
        if (element == null)
            continue;
        var text = element as Text;
        if (text != null && !string.IsNullOrEmpty(text.Value))
        {
            var token = new SimpleToken(text.Value);
            token.Span = new SegmentRange(run, 0, text.Value.Length -1);
            tokens.Add(token);
            run++;
        }
    }
    return tokens;
}

private List<SegmentRange> CollectConcordanceMatchRanges(Segment segment, string searchString)
{
    var concordanceMatchRanges = new List<SegmentRange>();
    int searchLength = searchString.Length;
    int run = 0;
    foreach (var element in segment.Elements)
    {
        var text = element as Text;
        if (text != null && !string.IsNullOrEmpty(text.Value))
        {
            int index = text.Value.IndexOf(searchString, StringComparison.OrdinalIgnoreCase);
            while (index >= 0 && index < text.Value.Length )
            {
                var segmentRange = new SegmentRange(run, index, index + searchLength -1);
                concordanceMatchRanges.Add(segmentRange);

                index += searchLength;
                if (index < text.Value.Length)
                    index = text.Value.IndexOf(searchString, index, StringComparison.OrdinalIgnoreCase);
            }
            run ++;
        }
    }
    return concordanceMatchRanges;
}

Then I test the concordance with this translation provider. I can get the result as shown in the screenshot:

Trados Studio Developers > Studio Developers Q&A

Text Formatting/Highlighting in the Concordance Results window.