public class PdfVector
{
    [VectorStoreRecordKey]
    public ulong Key { get; set; }
 
    [VectorStoreRecordData]
    public string FileName { get; set; }
 
    [VectorStoreRecordData]
    public string Text { get; set; }
 
    [VectorStoreRecordVector(384, DistanceFunction.EuclideanDistance)]
    public ReadOnlyMemory<float> Vector { get; set; }
}

const string databasePath = "pdf_vectors.db";
if (!File.Exists(databasePath))
{
    File.Create(databasePath).Dispose();
}
 
using var connection = new SqliteConnection($"Data Source={databasePath};");
await connection.OpenAsync();
 
connection.EnableExtensions(true);
connection.LoadExtension("./extensions/vec0.dll"); // 確保 vec0.dll 在項(xiàng)目目錄

var vectorStore = new SqliteVectorStore(connection);
var pdfsCollection = vectorStore.GetCollection<ulong, PdfVector>("pdfs");
await pdfsCollection.CreateCollectionIfNotExistsAsync();

static List<string> ExtractPdfChunks(string filePath, int chunkSize = 500)
{
    var textBuilder = new StringBuilder();
 
    using var pdf = UglyToad.PdfPig.PdfDocument.Open(filePath);
    foreach (var page in pdf.GetPages())
        textBuilder.AppendLine(page.Text);
 
    string fullText = textBuilder.ToString();
 
    var chunks = new List<string>();
    for (int i = 0; i < fullText.Length; i += chunkSize)
    {
        int length = Math.Min(chunkSize, fullText.Length - i);
        chunks.Add(fullText.Substring(i, length));
    }
 
    return chunks;
}

IEmbeddingGenerator<string, Embedding<float>> generator =
    new OllamaEmbeddingGenerator(new Uri("http://localhost:11434/"), "all-minilm:latest");

ulong keyCounter = 0;
 
foreach (var file in pdfFiles)
{
    var chunks = ExtractPdfChunks(file);
 
    foreach (var chunk in chunks)
    {
        var vector = await generator.GenerateEmbeddingVectorAsync(chunk);
        var record = new PdfVector
        {
            Key = keyCounter++,
            FileName = Path.GetFileName(file),
            Text = chunk,
            Vector = vector
        };
        await pdfsCollection.UpsertAsync(record);
        Console.WriteLine($"Upserted chunk from {record.FileName}");
    }
}

Console.WriteLine("Enter your question:");
var query = Console.ReadLine();
var queryEmbedding = await generator.GenerateEmbeddingVectorAsync(query);
 
var searchOptions = new VectorSearchOptions
{
    Top = 3,
    VectorPropertyName = "Vector"
};
 
var results = await pdfsCollection.VectorizedSearchAsync(queryEmbedding, searchOptions);
 
await foreach (var result in results.Results)
{
    Console.WriteLine($"File: {result.Record.FileName}");
    Console.WriteLine($"Text: {result.Record.Text}");
    Console.WriteLine($"Score: {result.Score}");
    Console.WriteLine(new string('-', 50));
}