v2.2.0

CHANGELOG.md

@@ -5,6 +5,12 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## v2.2.0
+
+Add:
+
+- Added `BasicDiskVectorDatabase` to provide a basic automatically disk persistent vector database, and associated `BasicDiskVectorStore` and `BasicDiskVocabularyStore`.
+
 ## v2.1.3
 
 Add:

docs/docs/persistence/index.md

@@ -5,6 +5,8 @@ title: Data Persistence
 
 The `Build5Nines.SharpVector` library provides easy-to-use methods for saving a memory-based vector database to a file or stream and loading it again later. This is particularly useful for caching indexed content between runs, deploying pre-built vector stores, or shipping databases with your application.
 
+---
+
 ## :material-file: File Persistence
 
 `Build5Nines.SharpVector` supports persisting the vector database to a file.
@@ -51,6 +53,8 @@ vdb.LoadFromFile(filePath);
 await vdb.LoadFromFileAsync(filePath);
 ```
 
+---
+
 ## :material-file-move: Persist to Stream
 
 The underlying methods used by `SaveToFile` and `LoadFromFile` methods for serializing the vector database to a `Stream` are available to use directly. This provides support for reading/writing to `MemoryStream` (or other streams) if the vector database needs to be persisted to something other than the local file system.
@@ -92,3 +96,30 @@ vdb.DeserializeFromBinaryStream(stream);
 // deserialize asynchronously from JSON stream
 await vdb.DeserializeFromBinaryStreamAsync(stream);
 ```
+
+---
+
+## :material-file-database: BasicDiskVectorDatabase
+
+The `BasicDiskVectorDatabase` provides a basic vector database implementation that automatically stores the vector store and vocabulary store to disk. It's implmentation of vectorization is the same as the `BasicMemoryVectorDatabase`, but with the modification that it automatically persists the database to disk in the background to the specified folder path.
+
+Here's a basic example of using `BasicDiskVectorDatabase`:
+
+```csharp
+// specify the folder where to persist the database data on disk
+var vdb = new BasicDiskVectorDatabase("C:/data/content-db");
+foreach (var doc in documents)
+{
+    vdb.AddText(doc.Id, doc.Text);
+}
+
+var results = vdb.Search("some text");
+
+```
+
+### Tips
+
+- Prefer absolute paths for the storage folder in production services.
+- Place the folder on fast storage (SSD) for best indexing/query performance.
+- Avoid sharing the same folder across multiple processes concurrently.
+- Back up the folder regularly to preserve your vector store and vocabulary.

src/Build5Nines.SharpVector/BasicDiskMemoryVectorDatabaseBase.cs

@@ -0,0 +1,29 @@
+using Build5Nines.SharpVector.Id;
+using Build5Nines.SharpVector.Preprocessing;
+using Build5Nines.SharpVector.Vocabulary;
+using Build5Nines.SharpVector.Vectorization;
+using Build5Nines.SharpVector.VectorCompare;
+using Build5Nines.SharpVector.VectorStore;
+
+namespace Build5Nines.SharpVector;
+
+/// <summary>
+/// Base class for an on-disk vector database. Mirrors MemoryVectorDatabaseBase generic composition
+/// while using disk-backed stores for persistence.
+/// </summary>
+public abstract class BasicDiskMemoryVectorDatabaseBase<TId, TMetadata, TVectorStore, TVocabularyStore, TVocabularyKey, TVocabularyValue, TIdGenerator, TTextPreprocessor, TVectorizer, TVectorComparer>
+    : VectorDatabaseBase<TId, TMetadata, TVectorStore, TVocabularyStore, TVocabularyKey, TVocabularyValue, TIdGenerator, TTextPreprocessor, TVectorizer, TVectorComparer>
+    where TId : notnull
+    where TVocabularyKey : notnull
+    where TVocabularyValue : notnull
+    where TVectorStore : IVectorStoreWithVocabulary<TId, TMetadata, TVocabularyStore, TVocabularyKey, TVocabularyValue>
+    where TVocabularyStore : IVocabularyStore<TVocabularyKey, TVocabularyValue>
+    where TIdGenerator : IIdGenerator<TId>, new()
+    where TTextPreprocessor : ITextPreprocessor<TVocabularyKey>, new()
+    where TVectorizer : IVectorizer<TVocabularyKey, TVocabularyValue>, new()
+    where TVectorComparer : IVectorComparer, new()
+{
+    protected BasicDiskMemoryVectorDatabaseBase(TVectorStore vectorStore)
+        : base(vectorStore)
+    { }
+}

src/Build5Nines.SharpVector/BasicDiskVectorDatabase.cs

@@ -0,0 +1,47 @@
+using Build5Nines.SharpVector.Vocabulary;
+using Build5Nines.SharpVector.Id;
+using Build5Nines.SharpVector.Preprocessing;
+using Build5Nines.SharpVector.Vectorization;
+using Build5Nines.SharpVector.VectorCompare;
+using Build5Nines.SharpVector.VectorStore;
+
+namespace Build5Nines.SharpVector;
+
+/// <summary>
+/// A basic disk-backed vector database using Bag-of-Words, Cosine similarity,
+/// disk-backed vector store and vocabulary store. Uses int IDs and string metadata.
+/// </summary>
+public class BasicDiskVectorDatabase<TMetadata>
+     : BasicDiskMemoryVectorDatabaseBase<
+        int,
+        TMetadata,
+        BasicDiskVectorStore<int, TMetadata, BasicDiskVocabularyStore<string>, string, int>,
+        BasicDiskVocabularyStore<string>,
+        string, int,
+        IntIdGenerator,
+        BasicTextPreprocessor,
+        BagOfWordsVectorizer<string, int>,
+        CosineSimilarityVectorComparer
+        >, IMemoryVectorDatabase<int, TMetadata>, IVectorDatabase<int, TMetadata>
+{
+    public BasicDiskVectorDatabase(string rootPath)
+        : base(
+            new BasicDiskVectorStore<int, TMetadata, BasicDiskVocabularyStore<string>, string, int>(
+                rootPath,
+                new BasicDiskVocabularyStore<string>(rootPath)
+            )
+        )
+    { }
+
+    [Obsolete("Use DeserializeFromBinaryStreamAsync instead.")]
+    public override async Task DeserializeFromJsonStreamAsync(Stream stream)
+    {
+        await DeserializeFromBinaryStreamAsync(stream);
+    }
+
+    [Obsolete("Use DeserializeFromBinaryStream instead.")]
+    public override void DeserializeFromJsonStream(Stream stream)
+    {
+        DeserializeFromBinaryStream(stream);
+    }
+}

src/Build5Nines.SharpVector/Build5Nines.SharpVector.csproj

@@ -9,7 +9,7 @@
     <PackageId>Build5Nines.SharpVector</PackageId>
     <PackageProjectUrl>https://sharpvector.build5nines.com</PackageProjectUrl>
     <RepositoryUrl>https://github.com/Build5Nines/SharpVector</RepositoryUrl>
-    <Version>2.1.3</Version>
+    <Version>2.2.0</Version>
     <Description>Lightweight In-memory Vector Database to embed in any .NET Applications</Description>
     <Copyright>Copyright (c) 2025 Build5Nines LLC</Copyright>
     <PackageReadmeFile>README.md</PackageReadmeFile>