fix: address review feedback

Author: rebasedming

NUGET-README.md

@@ -1,6 +1,6 @@
 # ParadeDB for Entity Framework Core
 
-The official [Entity Framework Core](https://learn.microsoft.com/en-us/ef/core/) integration for [ParadeDB](https://paradedb.com), including first-class support for managing BM25 indexes and running queries using the full ParadeDB API. Follow the [getting started guide](https://docs.paradedb.com/documentation/getting-started/environment#entity-framework-core) to begin.
+The official [Entity Framework Core](https://learn.microsoft.com/en-us/ef/core/) integration for [ParadeDB](https://paradedb.com), including first-class support for managing BM25 indexes and running queries using the full ParadeDB API. The integration covers both full-text and vector search — vector search indexes pgvector `vector` columns directly, with no separate pgvector ORM plugin required. Follow the [getting started guide](https://docs.paradedb.com/documentation/getting-started/environment#entity-framework-core) to begin.
 
 ## Requirements & Compatibility
 
@@ -10,6 +10,7 @@ The official [Entity Framework Core](https://learn.microsoft.com/en-us/ef/core/)
 | EF Core    | 8.0+                          |
 | ParadeDB   | 0.25.0+                       |
 | PostgreSQL | 15+ (with ParadeDB extension) |
+| pgvector   | Required for vector search    |
 
 ## Examples
 

README.md

@@ -36,7 +36,7 @@
 
 ## ParadeDB for Entity Framework Core
 
-The official [Entity Framework Core](https://learn.microsoft.com/en-us/ef/core/) integration for [ParadeDB](https://paradedb.com) (powered by the [`pg_search`](https://github.com/paradedb/paradedb) Postgres extension), including first-class support for managing BM25 indexes and running queries using the full ParadeDB API. Follow the [getting started guide](https://docs.paradedb.com/documentation/getting-started/environment#entity-framework-core) to begin.
+The official [Entity Framework Core](https://learn.microsoft.com/en-us/ef/core/) integration for [ParadeDB](https://paradedb.com) (powered by the [`pg_search`](https://github.com/paradedb/paradedb) Postgres extension), including first-class support for managing BM25 indexes and running queries using the full ParadeDB API. The integration covers both full-text and vector search — vector search indexes pgvector `vector` columns directly, with no separate pgvector ORM plugin required. Follow the [getting started guide](https://docs.paradedb.com/documentation/getting-started/environment#entity-framework-core) to begin.
 
 ## Requirements & Compatibility
 
@@ -46,49 +46,7 @@ The official [Entity Framework Core](https://learn.microsoft.com/en-us/ef/core/)
 | EF Core    | 8.0+                          |
 | ParadeDB   | 0.25.0+                       |
 | PostgreSQL | 15+ (with ParadeDB extension) |
-
-## Vector Search
-
-ParadeDB indexes pgvector `vector` columns directly inside its BM25 index — no
-pgvector ORM plugin is required. Map a `float[]` property to a `vector(n)`
-column, pick a distance metric for the index, and order by the matching
-distance function:
-
-```csharp
-public class Item
-{
-    public int Id { get; set; }
-    public float[]? Embedding { get; set; }
-}
-
-// Model configuration
-modelBuilder.Entity<Item>(entity =>
-{
-    entity.Property(x => x.Embedding).HasVectorType(384);
-
-    entity
-        .HasParadeDbIndex("items_idx", x => x.Id)
-        .HasField(x => x.Embedding, VectorMetric.Cosine);
-});
-
-// Top-K query: a `@@@` predicate (`EF.Functions.All` for match-all here) and
-// a LIMIT (`Take`) are required for the index to serve the query
-float[] queryEmbedding = GetQueryEmbedding();
-
-var results = await db
-    .Items.Where(x => EF.Functions.All(x.Id))
-    .OrderBy(x => EF.Functions.CosineDistance(x.Embedding, queryEmbedding))
-    .Take(10)
-    .ToListAsync();
-```
-
-`VectorMetric.L2` (`<->`, the default), `VectorMetric.Cosine` (`<=>`), and
-`VectorMetric.InnerProduct` (`<#>`) map to the `vector_l2_ops`,
-`vector_cosine_ops`, and `vector_ip_ops` operator classes. The distance
-function used in `OrderBy` must match the metric of the index opclass —
-`L2Distance` with `L2`, `CosineDistance` with `Cosine`, and `InnerProduct`
-with `InnerProduct` — otherwise the query still returns correct results but
-falls back to a sequential scan instead of Top-K index pushdown.
+| pgvector   | Required for vector search    |
 
 ## Examples
 

examples/README.md

@@ -49,8 +49,8 @@ dotnet run --project examples/Examples.csproj -p:Example=HybridRrf
 
 ## Vector Search (`examples/VectorSearch`)
 
-Runs Top-K nearest-neighbor queries over a pgvector column with the native
-vector support, backed by a bm25 index when the server supports it.
+Runs Top-K nearest-neighbor queries over a pgvector column with ParadeDB
+vector support, backed by a paradedb index.
 
 ```bash
 dotnet run --project examples/Examples.csproj -p:Example=VectorSearch

examples/Shared/ExampleSetup.cs

@@ -46,6 +46,9 @@ await db.Database.ExecuteSqlRawAsync(
             "CALL paradedb.create_bm25_test_table(schema_name => 'public', table_name => 'mock_items')"
         );
         await db.Database.ExecuteSqlRawAsync("DROP INDEX IF EXISTS mock_items_bm25_idx");
+        // The VectorSearch example may have left its own index behind, and newer
+        // pg_search versions allow only one ParadeDB index per table
+        await db.Database.ExecuteSqlRawAsync("DROP INDEX IF EXISTS mock_items_vector_idx");
         await db.Database.ExecuteSqlRawAsync(
             """
             CREATE INDEX mock_items_bm25_idx ON mock_items
@@ -102,14 +105,18 @@ public static async Task SetupHybridAsync(DbContext db)
     {
         await SetupMockItemsAsync(db);
         await db.Database.ExecuteSqlRawAsync("CREATE EXTENSION IF NOT EXISTS vector");
+        // Newer pg_search versions create mock_items with an embedding column of a
+        // different dimension; the examples use 384-dimensional embeddings
         await db.Database.ExecuteSqlRawAsync(
             """
             DO $$
             BEGIN
-                IF NOT EXISTS (
-                    SELECT 1 FROM information_schema.columns
-                    WHERE table_name = 'mock_items' AND column_name = 'embedding'
-                ) THEN
+                IF (
+                    SELECT atttypmod FROM pg_attribute
+                    WHERE attrelid = 'mock_items'::regclass
+                      AND attname = 'embedding' AND NOT attisdropped
+                ) IS DISTINCT FROM 384 THEN
+                    ALTER TABLE mock_items DROP COLUMN IF EXISTS embedding;
                     ALTER TABLE mock_items ADD COLUMN embedding vector(384);
                 END IF;
             END $$;

examples/VectorSearch/Program.cs

@@ -19,6 +19,8 @@
 await ExampleSetup.SetupHybridAsync(dbContext);
 await EmbeddingLoader.LoadAsync(dbContext);
 
+// Newer pg_search versions allow only one ParadeDB index per table
+await dbContext.Database.ExecuteSqlRawAsync("DROP INDEX IF EXISTS mock_items_bm25_idx");
 await dbContext.Database.ExecuteSqlRawAsync("DROP INDEX IF EXISTS mock_items_vector_idx");
 await dbContext.Database.ExecuteSqlRawAsync(
     """

src/Internal/ParadeDbOptionsExtension.cs

@@ -13,7 +13,6 @@
 using Npgsql.EntityFrameworkCore.PostgreSQL.Infrastructure;
 #endif
 
-
 namespace ParadeDB.EntityFrameworkCore.Internal;
 
 internal sealed class ParadeDbOptionsExtension : IDbContextOptionsExtension